Early Plant Succession in Loblolly Pine Plantations as Affected by Vegetation Management

A common study design has been used at 13 locat ions across the South to examine loblol ly pine (Pinus taeda L.) plantations established using four vegetation control treatments after mechanical site preparat ion: (a) No Control , (b) Woody Control , (c) Herbaceous Control for 4 yr , and (d) Total Control . This research, the Compet i t ion Omission Moni toring Project (COMP), is moni toring both pine growth and plant succession. During the f irs t 8 yr, the cover of herbaceous components and prevalent genera, along with pine, woody (nonpine) , and total herbaceous cover were est imated annually in September. Stem numbers and heights of arborescent and nonarborescent woody species were measured during the f irs t 5 yr and yr 8 . There were 101 prevalent genera of herbaceous plants and 76 species/genera of woody plants present on the s tudy s i tes , wi th a core group common to most . Herbaceous cover was rapidly reestabl ished on No Control and Woody Control treatments , wi th greater than 80% cover in the f irs t year. Af ter the f irs t year, herbaceous cover s teadi ly decl ined on No Control plots and was sustained when woody plants were e l iminated. In general , grasses dominated the herbaceous layer (mainly Andropogon and Panicum spp.) with cover peaking in yr 4. Woody control increased the actual cover of both grasses andforbs, but only the relat ive proport ion offorbs, which peaked in yr 1-2. Woody control also increased the actual cover of vines and semiwoodies (mainly nontargeted Rubus spp.) by yr 6-8, but only the relative cover of semiwoodies. Development of the pine canopy cover was s imilar wi th woody and herbaceous control , but pine heights were greater wi th herbaceous control . Interestingly, herbaceous control did not increase total woody cover until year 8, but the proportion of arborescent tree to nonarborescent shrub cover was increased. Most arborescent species and rootstocks became established in the first year. South. J. Appl. For. 19(3): 109-126. Intensive forest management is often cited as a major conecosystems to increase the production of certain species tr ibutor to the loss of species from forest communit ies (Probst inevitably leads to shif ts in the relat ive abundance of coexistand Crow 1991, Salwasser 1990, Norse et al. 1986). Aling species (Hunter 1990, Westman 1990). A prevalent though change in the species composition of forests is a objective of vegetation management (or forest weed control) natural consequence of succession, the management of forest is to alter species composition and relative abundance to NOTE: This study was funded in part by the members of the Auburn University Silvicultural Herbicide Cooperative. Cooperators in this study include the authors and Bill Pope, Potlatch Corporation; Mickey Rachal, Scott Paper Company; Edward Daly and Charles Hollis, International Paper Company; William S. Garbett, Union Camp; Dan Mixson, DuPont; Rick Applegate and C.W. Mueller, Packaging Corp. of America; Don McMahone, Williamette Industries; James A. McGriff, Georgia-Pacific Corporation; and George Tiley, Cavenham Industries. Study establishment was greatly assisted by Steven A. Knowe, Oregon State University; Kenneth Xydias, Resource Management Services; Lee Atkins, Timberland Enterprises; and Richard D. Iverson, American Cyanamid. Special thanks to John Freeman, Auburn University Herbarium, for numerous species determinations. The contributions of Erwin Chambliss and Kelly Robinson were invaluable. Use of trade names is for reader’s information and does not constitute official endorsement or approval by the U.S. Department of Agriculture to the exclusion of any suitable product or process. Pesticides used improperly can be injurious to humans, animals, and plants. Remember to read the entire product label and use only according to label instructions. Store pesticides in original containers under lock and key out of the reach of children and animals and away from food and feed. SJAF 19(3) 1995 109 Reprinted from the Southern Journal of Applied Forestry, Vol. 19, No. 3, August 1995. favor crop trees. Although most vegetation management treatments accomplish this goal , the magnitude and direct ion of species changes are not always identical or predictable. Wide ranges in diversity effects have been observed as a result of vegetation management, and their magnitude is dependent on the intensity of biotic control and abiotic habitat manipulation (Robinson 1978, Conde et al. 1983a, 1983b, Zutter and Zedaker 1988, Hansen et al. 199 1). Diversity and species maintenance are coming under increased regulatory control on both public and private forestland (Salwasser 1990). In the future, increases in crop tree growth and yield may not only have to be suff icient to just i fy the expenditures for vegetation management, but may be weighed against changes in the amenity values of wildlife and noncrop species maintenance as well. As a result, there is an increasing need to understand the impacts of woody and herbaceous plant control on species abundance and relative dominance over t ime. To assess these impacts, as well as the influence of vegetat ion management on loblol ly pine growth, a group of investigators initiated a unified study in 1984, cal led the Compet i t ion Omiss ion Moni tor ing Project or COMP (Miller et al. 1987). A prior report in SJAF focused on the 5 yr pine response and relat ive competi t ion levels (Miller et al . 199 1). This report summarizes the first 8 yr of successional dynamics . An objective of COMP is to describe secondary plant succession as it is altered by the vegetation management treatments of complete woody control and complete herbaceous control as compared to no control and complete control. The 13 plantation sites in COMP established from Louisiana to Virginia (Figure 1) provide a limited but unique network of locations for documenting such trends within loblol ly pine plantat ions in the Southeast , The tes t t reatments Figure 1. COMP plantation study locations relative to physiographic provinces (map recently complied by J.H. Miller and KS. Robinson based on Landsat imagery and other sources). 110 SJAF 19(3) 1995 permit us to examine the development of herbaceous and woody associations independently and collectively. It is believed that an understanding of secondary succession in these extreme treatment s i tuat ions should establ ish bounds of possible trends for management areas where complete component control is not achieved. Herbicide treatments for woody and herbaceous plant control were applied in 1992 to about 552,000 ac of forest lands in the Southeast with increasing use projected (Fallis 1993). This is the first study to document early succession in plantations over a wide number of locations for the central portion of the southern forest region dominated by loblolly pine. Prior reports have focused on the longleaf pine type (Pessin 1933), Florida slash pine (Grelen 1962, Burns and Hebb 1972, Ballet al. 1981,Condeetal. 1983a, 1983b,Neary et al. 1990), Texas pine-hardwood forests (Stransky et al. 1986), and scattered reportings of wildlife forage production (Wolters and Schmidtling 1975, Lewis et al. 1984, Blake et al. 1987). Other investigations have also examined vegetative response in loblolly plantations due to vegetation management treatments at selected si tes (Zutter et al . 1987, Zutter and Zedaker 1988, Locasio et al. 199 1). Results in this report address the following practical questions about forest plant succession, soil exposure, and floristic richness in intensively managed loblolly pine planta t ions : l How do vegetation control treatments alter early loblolly pine canopy development? l How does herbaceous control influence hardwood and shrub development? l How does woody control influence herbaceous cover development? l To what degree and duration may areas be unoccupied or uncovered by vegetation following moderately intensive mechanical site preparation, like chopping and burning, and supplemental control t reatments? a How does woody plant control al ter herbaceous component dynamics? l What herbaceous plant genera occurred in these pine plantat ions and did woody control t reatments al ter their occurr e n c e ? l What are the early successional trends of the most common herbaceous plants in pine plantat ions? l What are the early establishment patterns of woody plants after moderately intensive mechanical site preparation, like chopping and burning? l How does herbaceous control alter woody plant development and dynamics? l What woody plant species occurred in these loblolly pine plantations and did herbaceous control treatments alter their occurrence? These are thought to be common quest ions asked by and to forest resource managers regarding the impacts of vegetat ion management on stand development and associated flora. These findings should also provide a knowledge base for more detailed studies of succession and species diversity in southern pine plantat ions . Methods Study locat ions ranged in la t i tude from 30.5”-37.2”N and longitude from 78.5”-93.O”W (Table I). Average annual precipitation ranged by location from 40-60 in. for the 8 yr studied, while March through November amounts ranged from 26-43 in. Frost-free days typically vary from 270 days in the south to 160 days in Virginia. The soils for the most part are Ultisols that are low in bases and have subsurface horizons with clay accumulations, interspersed with recent alluvium. Vegetation of the region has developed until recently (last 200 yr) with frequent burning and extensive cultivation for 8,000+ yr by Native Americans during an interglacial warming period (Bartram 1940, Cronin et al. 198 1, Van Lear and Waldrop 1989, Doolittle 1992). All study locations probably were cultivated in the past 200 yr. Detai led methods of COMP have been presented in previous reports by Miller and others (1987, 1991) and are only reviewed here with appropriate elaborations. Common criteria for site selection and a common study design (with some differences) were used at the I3 plantation locations, which accommodated pooling of data across all locations to study regional trends. Major site and treatment similarities were: l Most s tudy si tes were located on prevalent soi l types of the region. l Prior stand conditions, harvesting and site preparation methods were typical for the region and similar to each other. Roller-drum chopping and prescribed burning were used at 10 locations, while ei ther windrowing, rebedding, or complete biomass harvesting was used at the other three locat ions. All si te preparation was performed during the growing season before study establishment and usually incorporated burning. l A common pine plant ing densi ty of 538 trees/at was used, except at two locations (565 and 622 trees/at). General characteristics of individual study sites are presented in Table 1 and their locat ions relat ive to physiographic provinces are shown in Figure 1. Immediately before and after planting the following competition control treatments were imposed: 1. No Control. 2. Woody Control for 5 yr. 3. Herbaceous Control for 4 yr. 4. Total Control (both woody and herbaceous plant control combining 2 and 3). These four treatments yield vegetat ion si tuat ions that are the corner extremes of a response surface that encompasses most competition conditions common to young plantations. Treatments 1 and 2 were used to study herbaceous succession with and without the woody component , and similar ly, t reatments 1 and 3 were used to document woody plant succession with and without the herbaceous component. Semiwoody plants (e.g., Rubus spp.) were considered here as herbaceous. In this unique approach, secondary succession was being documented as selective control treatments were being applied, unlike traditional ecological studies following a discrete, singular disturbance. The four treatments were established at the 13 locations using a randomized complete block design with 4 replications, with 2 exceptions. A fifth block was added at the only Flatwoods Coastal Plain site near Pembroke (GA) and a completely random design was used at Bainbridge (GA). Treatment plots were 0.25 ac and measurement plots were 0.09 ac. Loblolly pines were double planted (12 in. apart) on a 9 x 9 ft spacing, except at two operationally planted locations (Table 1). After the first growing season, pines were randomly thinned to one per spot, which assured uniform pine densi t ies . Woody plant control after pine planting was achieved by using nonsoil active herbicides-Garlon (triclopyr) and Roundup (glyphosate). Applications were by directed basal and foliar sprays and basal wipes. These herbicides and methods minimized damage to nontarget herbaceous plants. The same methods were used to control volunteer shortleaf (Pinus echinatu Mill.) and loblolly pines on all plots. Virginia pines (Pinus virginiana Mill.) at Appomattox (VA) were not controlled due to their prevalence on regeneration si tes in the Piedmont of Virginia. Herbaceous control treatments relied mainly upon annual broadcast applications of the pre-emergent herbicide Oust (sulfometuron) and shielded directed sprays of Roundup. Pre-establishment screening trials identified Oust rates that resulted in minimal damage to planted conifers, hardwoods, and shrubs. Treatment impacts to nontarget plants decreased greatly during the first 3 yr as control conditions were reached and nontarget plants grew larger. Vine suppression was attempted at half the locations, mainly in the first 3 yr on Woody Control plots, using directed sprays. Vines were cut also from pines for accurate diameter measurements and to minimize pine canopy interference. Thus, vine development was suppressed, but only partially during the first 3 yr. Within each interior measurement plot, three 9 x 18 ft sample plots were systematically established, with the corners at pine planting spots-a 0.01 ac sample per 0.09 ac measurement plot, yielding a 12% sample. Annually in September of yr l-5 and in yr 8, all woody rootstocks taller than 0.5 ft were recorded by species (genus for some nonarborescents) and height classes. Rootstocks were those judged to originate from the same central root system with one or more stems. Height classes were delineated by 1 ft intervals up to 12 ft and then by 5 ft intervals. For cover estimates, the three 9 x 18 ft sample plots were halved to yield six 9 x 9 ft subplots per measurement plot. Annually in September for yr 1-8, cover was visually estimated within each subplot for the herbaceous life-forms and