Restoration of Longleaf Pine Ecosystems

Longleaf Pine Seedling Survival and the Role of Fire on a Flatwoods Site

Cohen, Susan1, John Kush2, and Kim Ludovici 1 1USDA Forest Service, Southern Research Station, 3041 Cornwallis Rd.
Research Triangle Park, NC 27709 and 2 School of Forestry and Wildlife Sciences, 108 M. White Smith Hall, Auburn University, AL 36849


Natural regeneration of longleaf pine (Pinus palustris Mill.) is one of the most important tools natural resource managers have at their disposal to regenerate existing longleaf pine stands. National Forests are going to longer rotations and contemplating uneven-aged management for longleaf pine. Several studies on excessively drained sandhill sites have indicated there is a tendency for longleaf to regenerate primarily in gaps. The more poorly drained flatwoods sites have been largely ignored in regeneration and gap studies yet they are generally productive and contain high numbers of rare species. A study on the eastern Coastal Plain of North Carolina in the Croatan National Forest was established to examine the regeneration issue and the impacts of fire. These sites, on Onslow soils, contain second-growth longleaf forests and have been winter burned once every two to four years for the last 20 years.

The frequent prescribed burning regime has produced an intact understory groundcover with high species diversity and strong competition for the sparse resources. Nine 0.28-hectare plots were established around gaps in 2001. The plots averaged 102.6 tress/hectare with a range from 56.6 to 198.8 tress/ha. Overstory basal area ranged from 4.6 to 10.4 square meters/ha with an average of 7.02 square meters/ha. Grass-stage seedlings/ha ranged from 100 to 7,657 with an average of 2,533 seedlings/ha. Seedlings out of the grass-stage but having a height < 1.37 m averaged 303 seedlings/hectare with a range of 28.6 to 1,357 seedlings/ha. The study sites were prescribed burned this past winter. The impacts of fire on seedling survival will be related to a number of environmental conditions that have been measured. Proper management of longleaf pine forests will enable public land managers to meet the multiple-use demands placed on that ecosystem, while preserving a rare community type.


Longleaf pine stands once dominated more than 60 million acres in the Southeastern US, and are characterized by an uneven-aged, open park-like stand structure. Historic activities such as naval stores production, species conversion and fire suppression, caused longleaf pine stands to become critically endangered communities of the southeastern USA. Aside from being the dominant tree species, longleaf pine is also considered a keystone species of this community type. Reestablishing and regenerating longleaf pine seedlings is a major priority for conserving and restoring longleaf pine forests throughout the region. Very little is known about longleaf pine regeneration on flatwoods sites. With National Forests going to longer rotations and contemplating uneven-aged management for longleaf pine, proper management of longleaf pine forests is needed by the National Forest Systems to meet the multiple-use demands placed on that ecosystem, while preserving a rare community type.

The Croatan National Forest, located on the North Carolina coast, contains 64,300 hectares between Morehead City and New Bern. There is only an estimated estimate 5,062 hectares of longleaf pine remaining (Outcalt and Sheffield, 1996) on the Croatan National Forest. Today, one of the primary goals of forest management on the Croatan is restoration and management of longleaf pine communities. The major tools in this effort are proper use of fire and utilization of natural regeneration.

Our ability to restore and manage fire-dependent ecosystems will depend on improving our understanding of multiple regeneration processes. While nursery production and plantation management have improved artificial longleaf pine seedling establishment, our understanding of natural regeneration processes has not. This study involves the collection of baseline information on stand and site characteristics. This information will be regressed against seedling and adult tree location to test site-specific impacts such as temperature, moisture, light and competition.

Study sites

Study sites are located on the Croatan National Forest, Carteret County, NC. All plots are on the Onslow soil series, a moderately to somewhat poorly drained, loamy sand (fine-loamy, siliceous, thermic Spodic Paleudults). This soil is highly acidic, and generally nutrient poor (Goodwin 1977). Annual precipitation in the region averages 1210 mm but extended droughts occur during the growing season. Mean annual temperature is 17oC with the coldest temperatures in January (0.5oC) and the warmest in July (32.9oC).

The study sites have a typical longleaf flatwoods/savanna vegetative structure, with a mature, uneven-aged overstory dominated by longleaf pine (scattered loblolly and pond pines are present), no midstory exists, and a mix of low-growing woody and herbaceous vegetation dominates the understory. Average stand age ranges from 70 to 100 years. For the past 20 years, these sites have been winter burned every 2-4 years. All sites have been burned within the last year. The sites contain similar dominant understory plants including: Gaylussacia spp., Vaccinium spp., Ilex spp., Persea borbonia, Magnolia virginiana, Aristida stricta, Andropogon spp., Pteridium aquilinum, Eupotorium spp.

Objectives and Hypotheses

The proposed study addresses conditions of regeneration within gaps in natural longleaf pine stands, and intends to fill a void in regeneration information on moderate to poorly drained soils of the eastern North Carolina Coastal Plain. This study provides critical information for the Croatan National Forest to use in its management and restoration of longleaf communities. We plan to develop a monitoring program to examine relationships among longleaf pine seed crop, seedling/sapling mortality, and subsequent growth to edaphic conditions, stand conditions, and prescribed fire.

The significance of longleaf pine cone crops:

Natural regeneration is the most important tool natural resource managers have at their disposal to ecologically and economically regenerate existing longleaf pine stands. However, adequate cone crops for natural regeneration typically occur every 5-7 years, and often longer (Wahlenberg, 1946; Maki 1952) on the Atlantic Coast.

One of the major concerns in longleaf pine management is seed production. Compared to the other southern pines, longleaf is a sporadic seed producer. Wahlenberg (1946) noted that good seed crops might occur every 5 to 7 years. Maki (1952) reported heavy seed crops might occur over much of the longleaf range once in 8 to 10 years.

For successful regeneration, the minimum size of a cone crop is considered to be 1,853cones/hectare or roughly 30 cones per tree (Boyer and White 1989). In the past 30 years, 5 of the 8 cone crops considered adequate for natural regeneration have occurred since 1990 (Boyer 1998). The 1996-longleaf seed crop was one of those “much-anticipated” region-wide seed crops. Whether the interest is natural or artificial regeneration, it is important to know when to expect a bountiful seed crop. With only been 5 crops (50 cones/tree) considered adequate for regeneration, following what happens to the regeneration is critical to the management of longleaf pine.


A. Nine 0.28-hectare plots have been established at several locations on the Croatan National Forest.

B. Monitoring seedling crops: One-half of each plot was randomly selected and all seedlings with a DBH less than 1 cm were stem mapped from plot center. Seedling mortality has been followed through one growing season and one dormant season burn.

C. The longleaf pine overstory has been stem mapped and data recorded includes diameter at breast height (1.37 m; DBH), crown class, and crown and total height.

D. Longleaf pine flower and cone counts were made in 2000 and 2001 on a subset of trees on each plot.

E. Eleven 0.25 m2 litter traps have been set up on each plot and litter is being collected monthly. This will provide an estimate of productivity and fuel loads.

F. Soil temperature, soil moisture, and light will be measured/collected along transects monthly.

G. Soil cores have been taken every 10 cm to a depth of 30 cm, roots were sieved, and a subsample of soil was collected for C/N and organic matter content.

H. Root biomass is being determined by depth and separated into woody and non-woody components.

I. Vegetation surveys have been conducted and percent cover of understory plants mapped.

First-Year Findings

Plots have a typical flatwoods/savanna type structure: the overstory is dominated by longleaf pine with a few scattered loblolly and pond pines; there is no midstory, and a diverse understory.

The dominant herbaceous species is Aristida stricta on all plots. Other important herb/forb/grass species include: Polygala lutea, Pteridium aquilinum, Osmunda cinnamomea, Dichanthelium spp., Eupatorium spp., Xyris caroliniana and ambigua, Solidago odora, Aster spp., Rhexia spp., Aletris farinosa, Lysimachia quadrifolia, Hypericum spp., Drosera brevifolia and capillaris, Sarracenia purpurea and flava, Arundinaria gigantea, Sabatia spp., Platanthera spp., Calopogon spp., Cleistes bifaria, and Andropogon spp.

The dominant woody species are Vaccinium tenellum, Gaylussacia frondosa, Ilex glabra, Gaylussacia dumosa, Vaccinium crassifolium, Liquidambar styraciflua, Aronia arbutifolia, Acer rubrum, Myrica cerifera, Cyrilla racemiflora, Clethera alnifolia, and Lyonia spp.

Longleaf Pine Overstory

The plots averaged 102.6 tress/hectare with a range from 56.6 to 198.8 tress/hectare. However, they averaged only 77.0 trees/hectare that were large enough to be cone-bearing trees.

The average DBH was 27.0 cm with a range from 5.5 to 55.4 cm. Average height was 17.0 m with a range of 5.5 to 27.4 m. Basal area ranged from 4.6 to 10.4 square meters/hectare with an average of 7.02 square meters/hectare.

Description of gaps within plots

  • Gaps were mostly elliptical in nature and gap size averaged 0.09 hectares with a range of 0.01 to .16 hectares.
  • 56% of the gaps were oriented in a north-south direction.
  • The 2000 cone count found 26.5 cones/tree but the 2001 count had less than 2 cones/tree.
  • Grass-stage seedlings/hectare ranged from 100 seedlings/hectare to 7,657 with an average of 2,533 seedlings/hectare .
  • Seedlings out of the grass-stage but having a height < 1.37 m averaged 303 seedlings/hectare with a range of 28.6 to 1,357 seedlings/hectare

The remainder of the data is currently being processed and will be made available upon completion.

Works Cited

Boyer, W.D. 1973. Air temperature, heat sums, and pollen shedding phenology of longleaf pine. Ecology 54(2):420-426.

Boyer, W. D. 1998. Long-term changes in flowering and one production by longleaf pine. In: Proceedings of the Ninth Biennial Southern Silvicultural Research Conference, T.A. Waldrop (ed.), USDA Forest Service, Southern Research Station, General Technical Report-20, pages 92-98.

Boyer, W.D., and J.B. White. 1990. Natural regeneration of longleaf pine. In: Proceedings of the symposium on the management of longleaf pine, R.M. Farrar, Jr. (ed.), USDA Forest Service, Gen. Tech. Rep. SO-75, pages 94-113.

Gemmer, E.W., T.E. Maki, and R.A. Chapman. 1940. Ecological aspects of longleaf pine regeneration in south Mississippi. Ecological Monographs 21:75-86.

Goodwin, R.A. 1977. Soil survey of Carteret County, North Carolina. USDA Soil Service Conservation.

Maki, T.E. 1952. Local longleaf seed years. Journal of Forestry. 50(4):321-322.

Outcalt, K.W., and R.M. Sheffield. 1996. The longleaf pine forest: trends and current conditions. USDA Forest Service Resource Bulletin SRS-9. 23 p.

Wahlenberg, W.G. 1946. Longleaf pine: Its use, ecology, regeneration, protection, growth, and management. Charles Lathrop Pack Forestry Foundation with the USDA Forest Service, 429 p.