Managing Leaf Area for Maximum Volume Production in a Loblolly Pine Plantation

To manage loblolly pine (Pinus faeda L.) stands for maximum volume growth, the relationships between VOh’ne growth and leaf area at the tree and stand level under different cultural practices (thinning and fertilization) were examined. Forty-eight trees were harvested in 1995, six years after treatment, for individual tree measurements, and 336 standing trees were used for stand measurements each year from 1991 to 1994. Thinning significantly Increased annual ring width, tree leaf area, and tree volume growth during the six years following treatment, but reduced stand leaf area index and stand volume growth. Fertilization increased leaf area and volume growth at both the tree and the stand level, but significant tree level effects were only apparent during the first three years following treatment. The combination of thinning and fertilization was the optimum overall treatment. The relationship between volume growth and leaf area was linear and positive at the both of tree and stand level across treatments. INTRODUCTION Studies of leaf area and growth dynamics of loblolly pine stands in response to silvicultural practices are currently in progress as part of the USDA Forest Service, Southern Global Change Program. Leaf area is a key measurement variable in ecophysiology studies of both individual trees and forest stands, because it reflects the amounts of energy and material exchange between the forest canopy and the atmosphere. The leaf area of tree crowns and stands has a direct effect on photosynthetic capacity by affecting the surface area for carbon fixation, and an indirect effect on photosynthetic capacity by influencing the radiation, temperature, water vapor, wind, and carbon exchange within the canopy (Drew and Running 1975, Gholz and others 1991). Numerous studies have established a positive relationship between leaf area and forest productivity (Brix 1983, Shelbume and others 1993, Vose and Allen 1988). Silvicultural practices such as thinning and fertilization may increase leaf area, above-ground tree growth, and root growth in many conifers (Brix 1981, Dougherty and others 1995, Haywood 1994, Sword and others 1996 and 1998, Vose and Allen 1988). Brix (1981) examined the effects of thinning and fertilization on annual leaf biomass and total leaf biomass per tree in Douglas%. He found that thinning and fertilization increased both annual leaf biomass and total leaf biomass. Increases in annual leaf biomass peaked 2-3 years after fertilization, while the greatest differences in total leaf biomass were not found until 4-7 years after fertilization alone. Short-term fertilization responses in loblolly pine leaf area have also been reported (Gillespie and others 1994, Vose 1988). Vose and Allen (1988) found that fertilization increased the leaf area index (LAI) of loblolly stands 2 years after treatment in nutrient-limited stands and that LAI varied with stand density. Stemwood volume growth was positively and linearly related to LAI across treatments and stands. However, the long-ten effects of thinning and fertilization on loblolly pine leaf area and the relationship between leaf area and volume growth at tree and stand levels have not been reported. The objectives of this study were to (1) identify impacts of thinning and fertilization on tree volume growth; (2) assess the cultural practice effects on leaf area of individual trees and stands; and (3) determine the relationship between leaf area and volume growth. MATERIALS AND METHODS Study Site The loblolly pine (Pinus taeda L.) plantation in this study is located on the Palustris Experimental Forest in central Louisiana (31°07’N, 93’17’W). It was established in May 1981 when 14-week-old container-grown loblolly pine seedlings were planted at a spacing of 1.83 m by 1.83 m (2990 trees per ha). Twelve research plots, uniform in terms of tree size and spacing, were established within the plantation in the fall of 1988. Each plot was 23.8 m by 23.8 m (0.06 ha) and consisted of 13 rows of 13 trees. Two levels each of thinning and fertilization treatments were randomly assigned to 12 plots in a two by two factorial design resulting in four treatment combinations (thinned-fertilized, thinnedunfertilized, unthinned-fertilized, and unthinned-unfertilized) with three replications of each. On the thinned plots, 75 percent of the trees were removed in November 1988 by harvesting every other row of trees and every other tree in the remaining rows to produce a density of 731 trees per hectare. On the fertilized plots, diammonium phosphate at 744 kg per hectare (150 kg P + 134 kg N ha”) was broadcast in April 1989. The soil is a Beauregard silt loam (fine-silty, siliceous, thermic, Plinthaquic Paleudults). Soil drainage was adequate and slope was sufficient to prevent water from standing on the site (Haywood 1994). Sampling and Measurements Individual tree measurements-In early spring 1995,48 trees were harvested from 12 plots (two dominant or codominant trees and two intermediate trees (lower in basal area) from each plot). Immediately after felling each sample tree, the height, live crown length and diameter at breast height (d.b.h.) were measured. Three disks were cut from each tree (one at breast height, one at the base of the live crown and one at the middle of the crown). The disks were placed in plastic bags, taken to the laboratory, and placed in cold storage until they were analyzed. The ring width and ’ Paper presented at the Tenth Biennial Southern Silvicultural Research Conference, Shreveport, LA, February 16-16, 1999. * Graduate Research Assistant, Associate Professor, Professor, and Post-Doctoral Researcher, School of Forestry, Wildlife, and Fisheries, Louisiana Agricultural Experimental Station, Louisiana State University Agricultural Center, Baton Rouge, LA; Research Forester, USDA Forest Service, Southern Research Station, Pineville, LA, respectively. inside bark diameter for each year between 1988 to 1994 were measured to the nearest 0.1 mm in two vertical directions on the disk and then averaged. Specific leaf area (projected leaf area per unit dry weight) was calculated from 98 sample shoots per treatment combination (Vu, 1998). Projected leaf area was measured to 0.01 cm* using a LI-3000 Leaf Area Meter (LI-COR Inc., Lincoln, NE USA). Needles of each shoot were oven-dried (60 OC) to a constant mass and weighed to 0.01 g. Mean tree projected leaf area was calculated based on needlefall dry weight, specific leaf area, and number of trees per plot. Tree heights in the previous years were predicted from d.b.h. using the following equation: In(H)=a,+a,(l/D) (1) where H = total height, D = diameter at breast height, and aoI al = coefficients. Tree volumes in each year were calculated separately for thinned and unthinned treatments based on Baldwin and Feduccia’s volume equations (1987). Stand measurements-The measurement plot was the interior portion of each treatment plot originally occupied by the central 7 rows of 7 trees each (0.015 ha). A total of 338 trees from the measurement plots were used to assess d.b.h. and 144 of these trees were used to measure total tree heights at the end of each growing season from 1991 through 1994. Stand volumes for each treatment and each year were obtained by summing tree volumes and expanding to a per hectare basis. Needlefall was collected, oven-dried (60 “C) to a constant mass, and weighed each week from four litterfall traps per plot (i.e. 12 traps per treatment combination) in 1993, 1994, 1995, and 1998. Each trap was 0.92 m’. Litterfall dry weights were expanded to the plot level. Total leaf area per year and plot was estimated based on specific leaf area and two years of needlefall dry weights per plot (needlefall per year was summed from April to the following March) (Dougherty and others 1995, Vose and Allen 1988). Leaf area index was calculated from total leaf area per plot divided by plot ground area. Statistical Analysis The statistical significance of the thinning and fertilization main effects and the thinning by fertilization interaction effect was determined using analysis of variance (Statistical Analysis System, SAS V 6.12, SAS Institute Inc., Cray. NC, USA). Treatment effects on d.b.h., tree volume, stand volume, leaf area, and leaf area index in different years were analyzed by a two by two factorial in a completely random design. Treatment effects and their interaction with year on annual ring width, annual increment for diameter, tree height, tree volume, and stand volume were detem-rined by a two by two factorial repeated measurements. Main and interaction effects were tested for statistical significance at the 0.1 probability level due to the low number of sample trees and large natural variation in forests. If the interaction of thinning by fertilization was significant, the simple effects were tested (Stehman and Meredith 1995) using adjusted Tukey test (Geaghan, J.P., personal communication, Department of Experimental Statistics, Louisiana State University). RESULTS AND DISCUSSION Individual Tree Level Thinning and fertilization significantly increased d.b.h. (fig. IA). The effects of thinning or fertilization on annual ring width significantly interacted with year. In 1988 before treatment, annual ring width was uniform. After treatment beginning in 1989 and continuing through 1994, thinning significantly increased annual ring width by 113, 191, 236, 145,165, and 107 percent by year, respectively, compared to the unthinned treatment (fig. 1 B). Fertilization significantly interacted with thinning during the first two years after treatment (table 1). Fertilization increased annual ring widths by 36 and 50 percent within the thinned plots in 1989 and 1990, but had no significant effect on the unthinned plot tree ring widths. In 1991, fertilization increased annual ring width by 35 percent. After 1992, the fourth year after treatment, fertilization effects on annual ring width were not