1994 Proceedings
North American Conference on Savannas and Barrens

Savannas dominated by a sparse canopy of black cherry (Prunus serotina) and (or) red maple (Acer rubrum) develop sporadically in stream valleys, at drainage basin heads, and on high plateaus in the Allegheny Plateau Physiographic Province of northwestern Pennsylvania and southwestern New York. Locally called orchard stands, savannas were an uncommon component of the presettlement hemlock-white pine-northern hardwood forest that dominated the region. They are primarily anthropogenic in origin having arisen following the extensive logging and subsequent hot slash fires that occurred at the turn of the century (Hough 1945; Marquis 1975; Horsley 1985; Whitney 1990). In the Allegheny National Forest alone, it is estimated that anthropogenic savannas and treeless areas currently comprise ca. 10,000 ha, ranging in size from 5 to 200 ha (Allegheny National Forest 1986).

Allegheny Plateau savannas are maintained by a complex of biotic and abiotic factors that limit woody plant recruitment including heavy browsing of seedlings by deer and small mammals, seasonal microclimatic extremes, spatially limited seed sources, intense competition for light and soil resources with herbaceous plants, and allelopathy (Hough 1945, 1949; Horsley 1977, 1979, 1987). Past research on Allegheny Plateau savannas has primarily focused on restoration of site productivity and enhancement of Allegheny hardwood regeneration, particularly on high plateaus (Horsley 1985). The structure and diversity of plant communities associated with Allegheny Plateau savannas, particularly those that occur in larger stream valleys, is poorly known. This abstract provides a preliminary summary of on-going work on the vegetational diversity of a stream valley savanna in northwestern Pennsylvania, emphasizing composition of the woody vegetation and vascular plant diversity.

STUDY AREA

Studies were conducted in a stream valley savanna along Salmon Creek, Allegheny National Forest (Forest County, Pennsylvania), as part of a larger study of riparian community structure in the unglaciated Allegheny Plateau. The site sampled was ca. 15 ha in area and probably arose from large-scale logging that occurred in the Salmon Creek Valley during the late 1800s (Casler 1976). The site was fairly level but microtopography within the savanna was diverse with numerous small, dry ridges alternating with low, wet depressions.

METHODS

The point-quarter method was used to sample trees > 10.0 cm dbh. A total of 30 points was systematically placed along three randomly located transects within the savanna. For sampling and monitoring ground vegetation, twenty-four permanent 2 x 2 m plots were established along the transects: 12 plots were established in low, wet sites and 12 plots in higher, dry sites. Woody plant seedlings (stems < 1 m in height) were sampled in the 2 x 2 m plots; herbaceous plants were sampled in a 1.0 m2 plot nested randomly in each larger plot. Density, basal area and frequency were determined for all trees. Importance values for trees were calculated as relative density + relative basal area + relative frequency/3 (Curtis and McIntosh 1951). Importance values for woody plant seedlings were calculated as (relative density + relative frequency)/2. Small trees (stems > 2.5 cm but < 10 cm dbh) and saplings (stems < 2.5 cm dbh and > 1.0 m in height) were absent from the site. The area was sampled in July 1993. Nomenclature follows Gleason and Cronquist (1991).

RESULTS

Salmon Creek savanna had a canopy cover 31.9 %. The density of trees (individuals > 10.0 cm dbh) averaged 54.7 stems/ha with a stand basal area of 7.1 m2/ha. Canopy vegetation was composed of six species and dominated by Prunus serotina, Populus tremuloides and Amelanchier arborea (Table 1). Carpinus caroliniana, Acer rubrum and Tsuga canadensis were minor canopy components.

Eleven woody plant species occurred in the seedling layer (Table 2). Dominant species were Rubus allegheniensis, A. arborea and P. serotina. The herbaceous vascular flora consisted of 41 species in two ecological groups (Table 3): 1) species of dry to mesic microsites (18 species), and 2) species of wet microsites (23 species; some sedge and grass species are yet to be determined). Diversity of herb species was substantially higher than that recorded for plateau savannas; only 4 dry microsite species were common to the Salmon Creek savanna and plateau savannas (Table 3).

DISCUSSION

Preliminary results indicate that stream valley savannas of the Allegheny Plateau can support a diverse herbaceous flora consisting of a mix of upland and wetland species, a marked contrast to the low plant diversity typical of upland savannas (Horsley 1985). The greater herb species diversity at Salmon Creek probably is due to greater site heterogeneity especially the presence of wet and mesic microsites which do not occur in most upland savannas. We expect the Salmon Creek savanna species list to increase with expanded sampling across seasons.

Anthropogenic savannas are often considered wastelands incapable of timber production. Research to date indicates that restoration of savannas to hardwood forests is exceedingly difficult owing to numerous factors that limit growth and recruitment of desirable hardwoods (Horsley 1977, 1985, 1987). In light of this information, Allegheny National Forest (1986) has proclaimed savannas as permanent wildlife openings. Studies addressing the potential contribution of anthropogenic savannas to local and regional biodiversity of the Allegheny Plateau are sorely needed. For example, anthropogenic savannas may be seasonally important activity sites for some birds, reptiles, amphibians and lepidopterans. The biodiversity value of anthropogenic savannas may far outweigh their diminished role in timber production on the Allegheny Plateau.

ACKNOWLEDGMENTS

Melanie Haggard, April Moore, Jon Settlemyer, Phil Smith and Gary Walters assisted with vegetation sampling. This study was supported in part by a participating agreement between the USDA Forest Service, Allegheny National Forest and Clarion University of Pennsylvania.

LITERATURE CITED

Allegheny National Forest. 1986. Land and resource management plan. USDA Forest Service. Warren, PA. 345 p.

Casler, W. C. 1976. Teddy Collins Empire: A Century of Logging in Forest County. Book No. 9. Logging Railroad Era of Lumbering in Pennsylvania. Lycoming Printing Co., Williamsport, PA.

Curtis, J. T. and R. P. McIntosh. 1951. An upland forest continuum in the prairie-forest border region of Wisconsin. Ecology 32:476-496.

Gleason, H. A. and A. Cronquist. 1991. Manual of Vascular Plants of Northeastern United States and Adjacent Canada. 2nd ed. The New York Botanical Garden. Bronx, NY. 910 p.

Horsley, S. B. 1977. Allelopathic inhibition of black cherry by fern, grass, goldenrod, and aster. Canadian Journal of Forest Research 7:205-216.

• 1979. Chemicals from herbaceous plants maintain forest openings. Pennsylvania Forests 68:12-13, 24.
• 1985. Reforestation of orchard stands and savannas on Pennsylvania's Allegheny Plateau. Northern Journal of Applied Forestry 2:22-26.
• 1987. Allelopathic interference with regeneration of the Allegheny hardwood forest. In: G. R. Waller, ed. Allelochemicals: Role in Agriculture and Forestry. American Chemical Society.

Hough, A. F. 1945. Frost pocket and other microclimates in forests of the northern Allegheny Plateau. Ecology 26:235-250.

• 1949. Deer and rabbit browsing and available winter forage in Allegheny hardwood forests. Journal of Wildlife Management 13:135-141.

Marquis, D. A. 1975. The Allegheny hardwood forests of Pennsylvania. USDA Forest Service General Technical Report NE-15. Northeastern Forest Experiment Station. Upper Darby, PA. 32 p.

Whitney, G. G. 1990. The history and status of the hemlock-hardwood forests of the Allegheny Plateau. Journal of Ecology 78:443-458.

Table 1. Importance values of large trees (stems > 10 cm dbh) in the Salmon Creek savanna. Headings are: RF = relative frequency, RD = relative density, RBA = relative basal area, and IV = importance value.