The Successional Status of Table Mountain Pine

1.0 Introduction

Table mountain pine (Pinus pungens Lamb.) is a genetically diverse species endemic to the Appalachian Mountains. The geographic range for table mountain pine extends from Pennsylvania to northern Georgia, while the elevational range varies. Table mountain pine has medium to thick bark, serotinous cones, self-pruning limbs, a deep rooting habit, and is pitch producing, all of which are characteristics of trees adapted to repeated occurrences of surface fires (Sutherland et al. 1993). This tree is a secondary pioneer species that will establish quickly on sites that have been disturbed, especially by fire. Thus, the species plays a major role in the regeneration of mountain forests after major fire occurrences (Zobel, 1969; Williams and Johnson 1990).

At the same time, table mountain pine has a lifespan of ca. 200 years (Zobel 1969), suggesting that the species not only helps to regenerate the forest after fire, but also becomes an integral facet of the forest community. Table mountain pine provides a unique habitat for many animal species. White-tailed deer, wild turkey, scarlet tanagers, and ruffed grouse, among other species, have all been observed in table mountain pine stands. The seeds of the tree provide food for many birds and small mammals; the mountain pine coneworm feeds only on the cones of table mountain pine. Additionally, its serotinous cones provide a food source that is available even when seed crops from other conifer species fail (Williams 1992).

1.1. Project Justification

Previous studies of table mountain pine have shown it to be a fire-dependent species (Zobel 1969; Williams et al. 1990; Williams and Johnson 1990; Sutherland et al. 1993; Waldrop and Brose 1999). Without fire, the reproductive success of table mountain pine is greatly reduced, while the forest slowly progresses into a stand of hardwood species. Due to this adjustment in forest composition and structure, table mountain pine may lose dominance, and perhaps be eliminated from the forest altogether. Hence, the existence of table mountain pine in late successional stands may be jeopardized by human alterations of the natural fire regime.

Our project is relevant to Task 1 of the Joint Fire Science Program (JFSP) RFP 2001-1. Table mountain pine is an ecologically valuable species found only in the Appalachian region. Previous studies have concluded that table mountain pine is a fire-dependent species; however, the site-specific fire history associated with table mountain pine stands remains unknown. It is imperative that forest managers understand the history of fire (including frequency, seasonality, and areal extent) in table mountain pine stands within Great Smoky Mountains National Park and surrounding areas in help ensure the survivability of the species.

Our proposed research will provide baseline information on the successional status of table mountain pine stands within the park and the fire history within those stands. Our findings will be relevant to Task 3 of the JFSP RFP 2001-3 by addressing local knowledge gaps that should prove beneficial to land management agencies. We will combine information from our dendroecological investigations of fire history with information on the current age structure of table mountain pine stands to provide. The dual-nature of our methods should provide a detailed summary of the current status of the age structure of table mountain pine populations, their relationships with past fire, and their prognosis for survival under the new human-altered fire regime. Additionally, we will collect information on forest composition and structure that will be valuable to managers as they begin to prescribe fire as a management tool to areas of the park. Our findings will allow managers to determine the optimum strategy for prescribing fire in table mountain pine stands and to determine the effectiveness of those prescribed burns by preserving data that may be eliminated by fire.

The majority of our study will be conducted within the boundaries of Great Smoky Mountains National Park, primarily because park personnel have documented the locations of table mountain pine stands with precise stand maps and detailed information about species composition and evidence of fire. Therefore, we will focus sampling efforts on these stands while sampling additional stands in surrounding national forests to provide an assessment of how fire varies spatially across the region. We have secured the endorsement and support from USDA Forest Service personnel at the Cherokee National Forest, the Nantahala National Forest, and the Bent Creek Experimental Forest in Asheville, North Carolina.

1.2 Project Objectives

Our project is designed to use dendroecological techniques to investigate both the age structure and fire history of table mountain pine populations within Great Smoky Mountains National Park and in the surrounding National Forests. Our main goal is to evaluate the critical factors necessary for the successful reintroduction of fire into those table mountain pine stands. Our study is retrospective in that we intend to establish reference conditions of forest dynamics as they existed prior to widespread fire exclusion and post-settlement human influences. In this sense, our study emphasizes that the past is the key to the present to better manage for the future. We have three primary objectives (1) to evaluate the current age structure of table mountain pine stands; (2) identify and characterize presettlement fire regimes (including wildfire frequency, seasonality, and area extent) in selected stands; and (3) combine this information to assess the current successional status of mid-elevation table mountain pine stands. Addressing these primary objectives will provide critical ecological information that will be invaluable to managers when reintroducing fire into forests where this pine species exists.

1.3 Background
1.31 Biogeography of Table Mountain pine

Table mountain pine was first collected by Michaux ca. 1794 near Tablerock Mountain in Burke County, North Carolina, and was later described by Lambert in 1803 (Zobel 1969; Sanders 1992). This species of pine is an Appalachian endemic found in small, dense, unevenly distributed stands throughout its range, which extends from Pennsylvania down the Appalachian Mountains to east Tennessee and northern Georgia. In its range, this species occupies xeric, south- and southwest-facing slopes, often on sites considered by foresters to be unfavorable for tree growth (Zobel 1969; Della-Bianca 1990; Sanders 1992; Sutherland et al. 1993; Turrill 1998). The range of table mountain pine falls exclusively within the ranges of two other yellow pines, pitch pine (Pinus rigida Mill.) and Virginia pine (Pinus virginiana Mill.). Table mountain pine is by far less common than the other two (Zobel 1969; Sanders 1992; Sutherland et al. 1993).

The altitudinal range of table mountain pine is rather wide, from approximately 305 to 1220 meters, with a few stands in Tennessee and North Carolina existing above 1220 meters. Stands of trees are most commonly found on steep, well-drained slopes and narrow ridges. This pine is associated with shallow, undeveloped soils that tend to be strongly acidic and highly infertile (Zobel 1969).

Table mountain pine is a shade-intolerant secondary pioneer species, highly adapted to both long- and short-interval fire regimes (Sutherland et al. 1993). It has serotinous cones that open and release seeds when heated. The seeds can remain viable for up to 11 years within the cone, although Barden (1978) found that approximately 40% of two-year-old seeds were released each year. In addition to serotinous cones, the pine can reproduce vegetatively from basal sprouts.

Table mountain pine has medium-thick bark, extremely long branches and deep roots. The long branches protect the typically thin soil from solar radiation that can quickly evaporate soil moisture, and the deep rooting habit anchors the pine firmly to the bedrock, while absorbing water and nutrients (Della-Bianca 1990). This species is commonly associated with pitch pine and chestnut oak (Quercus prinus L.) (Williams and Johnson 1990).

1.32 Previous Studies

Three major studies focused on the population dynamics of table mountain pine. The major objective of these studies was to identify any successional trends present in P. pungens populations of the southern Appalachian Mountains. Based on a suggestion by Whittaker (1956) that populations of drought resistant pine species (such as table mountain pine), will sustain themselves without fire on xeric south-facing slopes, Barden (1977) began a long-term study of table mountain pine populations on Looking Glass Rock in North Carolina.

It is uncommon for fires to be ignited by lightning in the southern Appalachians; the United States Department of Agriculture (1989) claims that less than three percent of all fires in the southern Appalachians are attributed to lightning strikes (in Sanders 1992). With only rare occurrences of natural fires, compounded by years of fire suppression in the southern Appalachians, Barden theorized that Whittaker's statement of pine persistence without fire must be true. Dendroecological techniques were used to determine the age structure of the table mountain pine populations on Looking Glass Rock. The results indicated continuous reproduction with a mortality rate of 50% since the last fire, which occurred in 1889 (Barden 1977).

Barden revisited the Looking Glass Rock study site in 1996, 20 years after his original table mountain pine study. He used the same techniques from the 1977 investigation; however, he arrived at different results. After twenty years, the populations of table mountain pine on his study site had a peak in the middle-age classes, with only a small percentage of trees in the younger age classes. According to Barden, this type of age structure is representative of aging or declining populations. He attributed this change to the droughts that occurred during the 1980s, with a possibility that an increase in global temperature might also have impacted the reproduction of table mountain pines (Barden 2000).

Williams and Johnson (1990) describe age distributions similar to that of the table mountain pine stands on Looking Glass Rock as typical of disturbance-dependent shade-intolerant pine species. Similar to Barden (1977), the purpose of their study was to test the suggestion by Whittaker (1959) that table mountain pine populations growing on xeric, south-facing sites can persist without fire. They also incorporated dendroecological techniques in their study of age structure. The age distribution of this species on three study sites differed from that on the sites studied by Barden (1977, 2000), because they found a bimodal distribution, with major peaks occurring from the 45- to 80-year age classes. Smaller peaks occurred for the 10-year age classes at all sites. Their results indicated that table mountain pine stands on Brush Mountain do not have an adequate percentage of trees in the younger age classes to promote persistence without the introduction of recurring fires (Williams and Johnson 1990).

During the Fourth Annual Dendroecological Fieldweek held in June 1993, Sutherland, Woodhouse, and Grissino-Mayer and others (1993) conducted a pilot project on the fire history of a stand of table mountain pines on Brush Mountain, Virginia, in the Jefferson National Forest. Though preliminary, this study served as the basis for the work we propose in this study. Most importantly, their findings demonstrated that dendrochronological study of wedges and sections from fire-scarred table mountain pine trees can be used to establish local records of fires in the eastern United States where reconstructions of past fires from the fire-scar record are virtually non-existent. In addition to their study using fire scars, the field week team extracted cores from individual pines to determine stand age structure, and made observations on the size structure of other trees in the stand. Results suggested a bimodal age distribution similar to that described by Williams and Johnson (1990) for other stands on Brush Mountain, with two major recruitment events apparently linked to two major fire events at the site. Thus, the pilot project by Sutherland et al. (1993) was also significant in providing the first evidence that recruitment phases postulated by others to be linked to fire were indeed temporally associated with past fires.

2.0 Materials and Methods
2.1 Study Areas

Our primary area of study will be concentrated in Great Smoky Mountains National Park (GSMNP), near the boundary between Tennessee and North Carolina. Additional sites will be located in the surrounding Cherokee and Nantahala National Forests, and possibly the Pisgah National Forest. Together, the forests in these areas comprise nearly 2 million acres, and are considered critical to the maintenance of biodiversity across the southeastern United States. The park and national forests are located near the most southern tip of the range of table mountain pine in eastern Tennessee and western North Carolina. The park consists of 800 square miles (2032 square kilometers) of land, of which 95% is forested. GSMNP is world-renowned for its high species diversity, and was visited by over 10 million people in 1999. It has been declared an International Biosphere Reserve by the United Nations (National Park Service 2000).

Though the local climate within GSMNP varies with aspect and altitude, the average climate for the southeastern United States is humid subtropical. The mean precipitation in GSMNP exceeds 80 inches (2030 millimeters). The annual temperature ranges from 53 to 88< F (12 to 32< C) in July to 19 to 51< F (-7 to 11< C) in January, with the average number of frost-free days ranging from 170 to 180 (Della-Bianca 1990; National Park Service 2000).

From the establishment of the national park and surrounding national forests until 1996, fire suppression was a major technique for management of its forests. The forests of the southern Appalachian Mountains are more susceptible to fire during the fall to spring months, when deciduous trees have dropped their leaves, thus adding new fuel to the forest floor (Sanders 1992). Very few fires occurred in the Park after the 1930s, and those that did were primarily ignited by human activity (Sanders 1992; Turrill 1998). The interval between fire recurrences was increased dramatically by the federal practice of fire suppression (Turrill 1998), which has had an impact on the reproductive ability of fire dependent species, such as table mountain pine.

In 1996, the National Park Service began to change the way fire was handled on public lands. In an attempt to prevent unintentional fires, prescribed (or controlled) burning was introduced to GSMNP to eliminate the fuel that had collected on the forest floor. Controlled burning not only prevents hazardous stand-eliminating fires, but also aids those species that depend upon fire for reproduction (NPS 2000). Prescribed burns, however, are used more as a management tool in the surrounding National Forests than in GSMNP. In 2000, controlled fires burned less than 1000 acres of forests in GSMNP, while personnel in Cherokee National Forest alone burned approximately 21,000 acres.

The specific study sites for our research will be selected after careful consideration of the local aspect, soil type and structure, slope, and elevation. We will use maps with detailed information created by NPS personnel during preliminary mapping of table mountain pine stands throughout much of the park beginning 1992. These maps are currently housed at the national Park Service's Twin Creeks Resource Center near Gatlinburg, Tennessee. Selected stands will represent a wide range of ecological and environmental factors that suggest the ability to support table mountain pines. Sites selection will be carefully coordinated with NPS and Forest Service personnel to ensure we provide adequate spatial resolution to infer the broad-scale effects of fire on table mountain pines and associated hardwoods.

2.2 Field Methods

Dendroecological techniques, similar to those used by Sutherland et al. (1993) in their Brush Mountain investigation, will be used in our project. To determine the age structure of the table mountain pine stands at our study sites, we will extract two cores from selected individuals at the base of the stem and parallel to the slope contour. We will sample enough individuals at each stand to ensure a representative age structure is attained. The age of seedlings and saplings will be determined in the field by counting terminal bud scars and branch nodes, when possible. This method has been used effectively to determine the age of table mountain pines too small to core (Williams and Johnson 1990). Relevant data (location, dbh, lean degree, lean direction, and crown condition) will be recorded on standard specimen forms.

We will analyze the forest structure at each study site by creating an inventory of all tree species present in each selected stand. In addition to an inventory of species, we will collect diameter at breast height (dbh), a measurement that will support the calculation of species importance values and tree density. Diameter of individuals less than 1.45 meters tall will not be measured. A sampling scheme similar to that used by Sutherland et al. (1993) will be used on our study sites to collect these data. Development of an understanding of the current forest structure at our study sites will provide information on the successional trajectory for each table mountain pine stand.

Fire-scarred living and dead table mountain pines will be located visually at each study site. Cross-sections or wedges will be collected from those individuals with multiple fire-scars, especially from any suitable standing snags or downed logs. In GSMNP, all cross sections will be collected using K-24 cross-cut hand saws only; no chain saw will be used. Under supervision, cross sections will be collected from selected sites in the surrounding national forests using both hand saws and a chain saw. Cross-sections will be labeled and wrapped with strapping tape to preserve sample integrity for transport back to the laboratory. Relevant information (location, dbh, lean degree, lean direction, and crown condition) will be recorded on standard specimen forms and will include drawings to assist cross-section reassembly.

2.3 Laboratory Methods

Increment cores will be glued to wooden core mounts, while cross-sections will be reassembled and mounted on plyboard when necessary. The samples will be sanded using increasingly finer grit sandpaper (40 grit through 320 grit) until the intra-annual ring detail is easily discernable under standard magnification. For cores that do not intersect the pith, the number of missing rings will be visually estimated from the curvature of adjacent rings using standard pith estimators. To accurately crossdate all rings from all samples, we will create skeleton plots that accentuate the narrow rings, culminating in a composite skeleton plot for the study area (Stokes and Smiley 1968). The composite skeleton plot, or master chronology, will be used to crossdate each sample and designate the exact year of formation for each tree ring. We will also identify easily recognizable signature patterns within the tree-ring structure to help identify and assign dates to the annual rings (Stokes and Smiley 1968; Grissino-Mayer 1995). When patterns are not discernable, we will input measured ring widths into the program COFECHA to statistically match the sample to the master chronology. COFECHA will also be used to evaluate the accuracy of the crossdating and measurement (Grissino-Mayer 1995).

Fire dates will be recorded and archived using FHX2 software (Grissino-Mayer 1995). In addition, the seasonality of the fire event will be determined and recorded by noting the approximate position of the fire scar within the annual ring (Baisan and Swetnam 1990): (1) scars in the early earlywood portion of the annual ring, (2) scars in the middle earlywood portion of the ring, (3) scars in the later earlywood portion, (4) scars in the latewood, (5) scars in the dormant position (between the latewood of one ring and the earlywood of the next ring), and (6) scars whose position can not be accurately determined. FHX2 software will be used to develop master fire charts (Dieterich 1980) depicting the temporal and spatial patterns of past fire at each site. We will use statistical analyses to develop estimates of the Weibull Median Fire Interval, Weibull Modal Fire Interval, Upper and Lower Exceedance Intervals, and Maximum Hazard Interval (Grissino-Mayer 1999). All statistical descriptors will help model presettlement fire regimes.

3.0 Project Duration

This research project is being undertaken to complete a required component of the Master's degree program being sought by Michael R. Armbrister in the Department of Geography at the University of Tennessee. We anticipate this project taking no more than one year to complete from date of award. If fieldwork commences in summer 2001, the analyses should be completed by 31 May 2002.

4.0 Deliverables
4.1 Tentative Schedule

July - August 2001: Collection of samples during summer on various field trips
August - December 2001: Development of fire history and age structure data
December 2001 - March 2002: Conduct all graphical and statistical analyses
March 2002 - May 2002: Write Final Report and deliver by 31 May 2002

4.2 Information and/or Products

At the end of the proposed project, we will deliver to personnel at Great Smoky Mountains National Park a detailed Final Report in the form of a Master's Thesis. The results will contain information about the fire history, age structure, and successional status of the table mountain pine stands. We will also provide scientifically sound interpretations concerning any evidence of direct and indirect effects of human disturbances on the health and status of table mountain pines gained from our analyses. We will also make recommendations concerning the possible restoration of fire in the park and its impacts on table mountain pines based on our interpretations. The Final Report will be distributed in digital form in standard formats (Word, RTF files), and will contain all data derived from our study in the form of appendices, and these data will also be made available in digital form.

4.3 Technology Transfer

We anticipate considerable interest in our study from federal, state, and local agencies, institutes, departments, scientists, students, and professionals involved in wildfire management and forest ecology. We have expertise in the development of online web-based teaching and research material, and intend to develop a specific site containing interconnected web pages dedicated to this study. For examples of teaching and research web pages already developed, please see:

Tree-Ring Web Pages: http://web.utk.edu/~grissino/
Online Soil Science Course: http://www.valdosta.edu/~grissino/geol3710/
Online Biogeography Course: http://www.valdosta.edu/~grissino/geog4900/

5.0 References

Baisan, C.H. and T.W. Swetnam. 1990. Fire history on a desert mountain range: Rincon Mountain Wilderness, Arizona, U.S.A. Canadian Journal of Forest Research 20: 1559-1569.

Barden, L.S. 1977. Self-maintaining populations of P. pungens Lam. in the Southern Appalachians. Castanea 42: 316-323.

Barden, L.S. 2000. Population maintenance of P. pungens Lam. after a century without fire. Natural Areas Journal 20: 227-233.

Della-Bianca L. 1990. P. pungens Lamb. In: R.M. Burns and B.H. Honkala, eds., Silvics of North America. Vol. 1 Conifers. USDA handbook 654.

Dieterich, J.H. 1980b. The composite fire interval - a tool for more accurate interpretation of fire history. In M.A. Stokes and J.H. Dieterich, eds., Proceedings of the Fire History Workshop. USDA Forest Service General Technical Report RM-81: 8-14.

Grissino-Mayer, H.D. 1995. Tree-ring reconstructions of climate and fire history at El Malpais National Monument, New Mexico. Doctoral Dissertation, The University of Arizona. 407 pp.

Grissino-Mayer, H.D. 1999. Modeling fire interval data from the American Southwest with the Weibull distribution. International Journal of Wildland Fire 9(1): 37-50.

Lorimer, C.G. 1980. Age structure and disturbance history of a southern Appalachian virgin forest. Ecology 61: 1169-1184.

National Park Service. 2000. Great Smoky Mountains National Park. http://www.nps.gov/grsm.

Sanders, G.L. 1992. The role of fire in the regeneration of Table Mountain pine in the southern Appalachian Mountains. Masters Thesis, The University of Tennessee, Knoxville.

Stokes, M.A. and T.L. Smiley. 1968. An Introduction to Tree-Ring Dating. The University of Chicago Press, Chicago.

Sutherland, E.K., H.D. Grissino-Mayer, C.A. Woodhouse, W.W. Covington, S. Horn, L. Huckaby, R. Kerr, J. Kush, M. Moore, and T. Plumb. 1993. Two centuries of fire in a southwestern Virginia Pinus pungens community. Paper presented at the IUFRO Conference on Inventory and Management in the Context of Catastrophic Events, University Park, PA. June 21-24.

Turrill, N. 1998. Using prescribed fire to regenerate Pinus echinata, P. pungens, and P. rigida communities in the southern Appalachian Mountains. Doctoral Dissertation, The University of Tennessee, Knoxville.

U. S. Department of Agriculture, Forest Service. 1989. Southern Region Annual Fire Report. Atlanta, GA.

Waldrop, T.A. and P.H. Brose. 1999. A comparison of fire intensity levels for stand replacement of table mountain pine. Forest Ecology and Management 113: 155-166.

Whittaker, R.H. 1956. Vegetation of the Great Smoky Mountains. Ecological Monographs 26: 1-80.

Williams, C.E. and W.C. Johnson. 1990. Age structure and the maintenance of P. pungens in pine-oak forests of southwestern Virginia. American Midland Naturalist 124: 130-141.

Williams, C.E., M.V. Lipscomb, and W.C. Johnson. 1990. Influence of leaf litter and soil moisture regime on early establishment of P. pungens. American Midland Naturalist 124: 142-151.

Williams, C.E. 1992. An Appalachian original. American Forests 98: 24-26.

Woods, F.W. and R.E. Shanks. 1959. Natural replacement of chestnut by other species in the Great Smoky Mountains National Park. Ecology 40: 349-361.

Zobel, D.B. 1969. Factors affecting the distribution of P. pungens, an Appalachian endemic. Ecological Monographs 39: 303-333.

6.0 Budget

A. Senior personnel PI, Co-PI’s, Faculty and Other Senior Associates: $5,000
B. Other personnel, Graduate Student Academic year, 740 hours X 12.00/hr Summer, 300 hours X $12.00/hr: $8,880 and $3,600
C. Fringe benefits: $1,400
D. Equipment: 0
E. Travel (domestic)
1. Field trips to GSMNP, Cherokee National Forest, and Nantahala National Forest, two individuals, gas, and per diem: 1,000
2. Required PI Workshop 500
F. Other direct costs
One 20" Haglof increment borer: $300
40 sanding belts @ $3/ea: $120
100 core mounts @ $2/ea: $200
Four boxes paper straws @ $25/box: $100
Miscellaneous field and shop supplies: $200
Graduate student tuition and fees: $3,460

Total Direct: $24,760
Total F&A at 45% MTDC: ($9,585)
Total F&A (15% required by JFSP): $3,195
Total Project Cost: $27,955

6.1 Budget Justification and Explanation
6.1.1 Salary

The University of Tennessee asks for salary for the Senior Principal Investigator on this project, Dr. Henri D. Grissino-Mayer (one/ninth annual salary = $5000). Salary is also requested for support of a Graduate Research Assistant, Michael Armbrister, for Summer 2001 at .75 FTE and during the Fall and Spring Semesters 2001-2001 at .50 FTE. UT Fringe benefits are calculated at 28% for faculty salary. No UT fringe benefits are associated with graduate student salaries.

6.1.2 Equipment

We require no additional specialized, expensive, permanent equipment for this project, using instead the equipment and resources already available at the Laboratory of Dendrochronology at the University of Tennessee. Such equipment includes personal computers, Velmex measuring systems, boom-arm stereozoom microscopes, and a Kodak imaging system.

6.1.3 Travel

We request funds to help offset costs associated with numerous field trips to Great Smoky Mountain National Park, the Cherokee National Forest, and the Nantahala National Forest. The funds include costs for gas and per diem at $30/day per person. We also request travel funds for the required PI workshop as mentioned in the RFP.

6.1.4 Other Direct Costs

We request funds for a limited number of heavy-use items, such as an increment borer and the assorted field and lab supplies required for this research. Also included are funds to cover tuition and fees for the graduate student.

6.1.5 Project Costs

Graduate student tuition and fees are not covered by Indirect Costs. Hence, the Total Indirect Costs are calculated on a revised Total Direct Cost value of ($24,760 - $3,460) = $21,300 X .15 = $3,195.

6.1.6 In-kind Contributions

In-kind contributions from the University of Tennessee total ($9,585 - $3,195 =) $6,390, reflected in the 30% reduction in Indirect Costs, from 45% asked for by the University to 15% required by the JFSP. In-kind contributions from the National Park Service include salary for the Park Ecologist (GS-11) at 0.1 work year = $4,200, and Biological Technician (GS-5) at 0.05 work year = $1,000.