Plant Communities in Selected Longleaf Pine Landscapes on the Catahoula Ranger District, Kisatchie National Forest, Louisiana

In Grant Parish, Louisiana, increases in overstory basal area, canopy cover, and development of understory woody plants reduced productivity Of herbaceous phts in longleaf pine (Pinus palustris Mill.) stands that were managed with fire. Still, the herbaceous plant community can reestablish itself on properly managed upland longleaf pine sites in the West Gulf Region. Management efforts were considered most successful where pinehill bluestem [Schizachyrium scoparium var. divsrgens (Hack.) Gould] is the dominant herbaceous plant. The lack of oak (Ouercus spp.) and hickory (Carya spp.) regeneration on more mesic sites was worrisome. Use of nested subplots was the best method for monitoring herbaceous vegetation. INTRODUCTION Fire was essential for the formation of many southern pine ecosystems. Today, failure to use prescribed fire in upland longleaf p ine landscapes resu l ts in encroachment by hardwood trees and shrubs and the loss of native pine and herbaceous vegeta t ion . For example , in A labama over 90 percent of the green biomass on the forest floor of young unburned longleaf pine stands is woody vegetation, while in periodically burned stands, less than 50 percent of the green biomass on the forest floor is woody vegetation (Boyer 1995). This woody vegetation can form a closed midstory that reduces species richness and productivity of the herbaceous p lan t communi ty (unpub l i shed f ie ld no tes) . Only about 5 percent of the pine forest lands in Louisiana is publicly owned (Vissage and others 1992). Given the limited acreage, it makes sense to manage these lands for rare and endangered species, old growth characteristics, and other things that may be beyond the management capability of private landowners. As public lands are managed fo r these o ther a t t r ibu tes , mon i to r ing becomes necessary for both legal and practical reasons. In January 1993, the Kisatchie National Forest and the Southern Research Station began monitoring the effects of operational-scale burning in longleaf pine forests on overstory and midstory trees and shrubs and understory vegetation. In addition, research studies on the Catahoula Ranger District (RD) have provided useful information about the effects of fire. We are reporting on the fire effects f rom opera t iona l -sca le burns done on two Ecosys tem Management Pro jec t (EMP) s i tes and w i l l compare those results to research findings. SITES All sites are on the Catahoula RD, Kisatchie National Forest, Grant Parish, LA. Elevations of the sites range from 53 to 76 meters (m). These sites are within the historical range of the upland longleaf pine forest type of the humid temperate, subtropical, outer coastal plain mixed forest, and are located in the coastal plains and flatwoods Western Gulf Ecoregion of the Southern United States (McNab and Avers 1994). The mean January and July temperatures are 10 and 28 “C, respectively (Louisiana Office of State Climatology 1995). Yearly precipitation averages 143 cen t imeters (cm) and g rowing-season precipitation averages 82 cm. The growing season is more than 200 days long: it usually begins before or during early March and ends because of dry weather in October. The two research sites are as follows: RESl : The site is a slightly sloping upland of Metcalf (Aquic Glossudal@ and Cadeville (Albaquic Hapludal@ very fine sandy loams. An existing stand [7,450 stems per hectare (ha)] of 6-year-old lobiolly pine (Pinus taeda L.) was clearcut and the debris burned before 0.093-ha study plots were established. For the next 11 years, woody vegeta t ion was cont ro l led by b ienn ia l burn ing and by severing of all woody and blackberry (Rubus spp.) stems over 1 m tall. Over the next decade, a pas tu re o f na t i ve woody and herbaceous p lan ts became reestablished. We are using data from plots burned biennially in early May from 1982 through 1992. RES2: The stand is a 17-ha longleaf pine shelterwood with reserves on a gently rolling upland of Ruston and Smithdale (Typic Paleudults) sandy loam soils. The shelterwood was established in 1968 when the initial preparatory cut left a residual basal area of 8.4 ma per ha. A seed-tree cut in 1975 left 6.9 m2 per ha of basal area. The seed trees have been reserved for the management o f red-cockaded woodpecker (Picoides borealis) habitat. The stand has been prescribed burned 11 times from 1969 through May 1993. Burns were during all seasons of the year. The two EMP sites are as follows: EMPL: The stand is a 188.ha lOngleaf pine forest on a Ruston, Smi thda le , and Malb is (P l in th ic Pa leudu l t ) sandy loam and loamy sand gently rolling upland. The two most recent prescribed burns were in 1990 and ’ Research Forester, Southern Research Station, USDA Forest Service, 2500 Shreveport Highway, Pineville, LA 71360; Research Forester. Southern Research Station, USDA Forest Service, Devall Dr., Auburn, AL 36649; and Silviculturist, USDA Forest Service, Kisatchie Natlonal Forest, 2500 Shreveport Highway, Pineville, LA 71360 (respectively). 8 6 Ill: aaldrop, Thorns A., ed. 1998. Pmceedhgs of the ninth biennial southern sllvicultural research conference; 1997 February 2527; Clemson, SC. Gen. Tech. Rep. SRS-20. Asheville, NC: U.S. Department of Agrlmdture, Forest Service, southern Resd Station. pm S--91 February 1993. Back ing and s t r iphead f i res were employed. EMP2: The stand is a 99-ha longleaf pine forest on a Ruston and Smi thda le sandy loam and loamy sand gently rolling upland. The three most recent prescribed burns were in 1990, July 1993, and May 1995. Sacking and flank fires were employed. PROCEDURES On RESI, total current-year herbaceous production was determined in February 1994 by c l ipp ing the aboveground fo l iage on 12 sys temat i ca l l y l oca ted 0 .22-m’ subp lo ts located within each 0.04.ha plot. Dry matter production (oven-dried at 80 “C for at least 24 hours) was determined after the samples were subdivided’ into six taxa: pinehiil b lues tem; o the r b lues tems-mos t l y b roomsedge (Andropogon virginicus L.), Elliott’s bluestem (A. elliottii Chapm.), big bluestem (A. gerarcfii Vitm.), and slender bluestem (S. tenerum Nees); longleaf uniola (Chasmanthium sessiliflorum (Poir) Yates]; other grassesmostly switchgrass (Panicurn virgatum L.), yellow indiangrass [Sorghastrum avenaceum (Michx.) Nash], low panicums (Dichanthelium spp), lovegrass (Eragrostis spp.), and threeawn (Aristida spp.).; grasslikes-mostly nutrush (Scleria spp.), sedge (Carex spp.), flatsedge (Cyperus spp.), spikesedge (Eleocharis spp.), rush (Juncus spp.), and beakrush (Rhynchospora spp.); and forbs. In March 1994, all woody, blackberry, and vine stems were counted and heights and crown spreads estimated on five systematically located 40-m’ subplots. On RES2, EMPl, and EMP2, O-04-ha plots were established for measuring heights and d.b.h. of the overstory and midstory trees. There were 16 plots on the uplands in RES2 and 10 plots on both EMPl and EMP2. Inventories were made in May 1996 on EMPI and EMP2 and in July 1996 on RES2. Within each 0.04-ha plot, five 4-m* subplots were systematically established for identifying and counting understory woody stems, blackberry stems, and vines ano for measuring heights and crown cover of the woody and blackberry stems. This brush was inventoried in April 1995 on EMP2, in August 1995 on EMPl, and in August 1996 on RES2. On RES2, five lines of 20 0.22-m* subplots were placed at equal distances apart across the site for inventorying herbaceous plant species. The lines were spaced 80 m apart and the subplots were spaced 16 m apart in each line. Only 86 of the 100 subplots fell on the upland soils. The remaining 14 subplots were in the Guyton (Typic Glossaqual ) dra inages and were not used. A l l herbaceous plants with root collars in the subplots were inventoried in July 1996. After the inventories, the herbaceous plants In the subplots were clipped to groundline to determine dry mat te r p roduc t ion . Twelve loo-point transacts were permanently located on both EMPl and EMP2. These were used to inventory herbaceous plants. Each transect was 30 m long and read ings were made every 30 cm. Read ings were made through a circular loop 2 cm in diameter. The loop was held about 30 cm from the eye and 60 cm above the ground. All herbaceous p lants seen through the loop were recorded. Readings were made in April 1995 on EMP2 and in August 1995 on EMPl. In September 1995, current-year herbaceous vegeta t ion was c l ipped to g round l ine w i th in seven 0.22-m* subplots adjacent to each transect to determine dry matter production on both EMPl and EMP2. RESULTS AND DISCUSSION Overstory and Midstory Vegetation On RES2, EMPl, and EMP2 total stocking and basal area ranged from 54 to 279 stems per ha and 8.0 to 24.4 m* per ha (table 1). Canopy cover was too sparse to measure accurately on RES2 but averaged 67 percent on EMPl and EMP2. Longleaf pine dominated the overstory on all sites Table l-Number and basal area of overstory and midstory trees and shrubs and the percentage of the stand in longleaf pine (Pinus palustris Mill.) Number o f s tems /ha Basal area (m*/ha) Standsa Total Longleaf p ine Total Longleaf p ine Longleaf p ine Tota l canopy cover _____PercentRES2she l te twood w i t h reserves 5 4 4 8 8.01 7.67 9 6 _ eb EMPIfo res t 2 7 9 124 24.36 19.85 8 1 7 7 EMP2fo res t 2 1 0 153 24.40 21.89 9 0 5 7 a RESI had no overstory or midstory vegetation. b No data for this sparse overstory. 8 7 f and made up from 81 to 96 percent of the total basal area. These three stands were classed as pure longleaf pine based on basa l a rea (Ford-Rober tson 1971) . More species of overstory and midstory trees and shrubs occurred on EMPl and EMP2 than on RES2. Species other than longleaf pine represented a greater portion of the stand basal area on EMPl and EMP2 than on RES2 (table 1). EMPl had nine common overstory and midstoty species: longleaf pine, mockernut hickory [Carya to