1994 Proceedings
North American Conference on Savannas and Barrens

Presettlement upland forests of the southern Illinois (Fralish et al. 1991), western Kentucky (Bryant and Martin 1988), western Tennessee (DeSelm 1988); Missouri (Ladd 1991), and specifically Land Between The Lakes (McCrain and Grubb 1988; Franklin 1994) are believed to have been more open than present forest communities. Following settlement, the suppression of fire permitted the open forests of presettlement Land Between The Lakes (LBL) into grow into a closed-canopy forest. Shading and the litter layer subsequently reduced the herbaceous vegetation.

The paper presents part of the results of a long term study of forest communities at LBL located in Kentucky and Tennessee. The objective here is to present community and environmental relationships on woodland communities that likely were once barrens and savanna-like.

STUDY AREA

Land Between The Lakes (LBL) is a 69,000 ha interfluve ca. 64 km long and up to 13km wide. The area is highly dissected with elevation ranging from 110 m to 190 m (Harris 1988; Franklin et al. 1993). The area is over 90% forested. The most common forest communities are chestnut oak (Quercus prinus), post oak (Quercus stellata), black oak (Quercus velutina) and white oak (Quercus alba).

The west side of the Tennessee Ridge, which bisects LBL, is generally dominated by coastal plain gravelly of Tertiary and Cretaceous age. The soil here is generally nutrient poor, drougthy, and unconsolidated, but large conglomerate boulders may be present. Limestone derived soil usually occurs on slopes and at lower elevations slopes particularly on the eastern and southern half of the property. A shallow loess layer frequently caps both the coastal plain gravel or limestone-derived soils on ridgetops and midslopes. The loessal soil may be droughty due to the presence of a shallow fragipan (Francis and Loftis 1977; Harris 1988; Fralish et al. 1993; Franklin et al. 1993).

METHODS

A study of forest communities at LBL was begun in 1987. Of the 133 1.0 - 2.0 ha study stands, 18 were dominated by chestnut oak and 14 by post oak. Stand selection criteria included an overstory of uniform age, no evidence of severe disturbance and no major canopy gaps. Minimum stand age was 80 yr. Stands were uniform with regard to topography. In each stand, we randomly located two points at least 27 m apart and 13 m from the stand boundary. Each point was used as the center of a 0.06 ha circular plot, in which all stems > 9 cm diameter at breast height (DBH) were measured. Saplings (1.0-9.0 cm DBH) were counted in four 0.006 ha circular plots located in cardinal directions 6.91 m from plot center. Seedlings (< 1.0 cm DBH) were counted in 0.003 ha plots nested within the 0.006 ha plots.

A soil profile description was made and samples taken for texture analysis. Aspect, elevation and slope also were recorded for each stand. Available water capacity was estimated from texture, percent stone, bulk density and depth of rooting, to a fragipan or to bedrock following the procedures of Fralish (1987).

Stands were separated into compositionally-stable and successional by comparing understory seedlings and saplings using percent similarity (PS = (2w/a + b). Stands with values less than 45 % were considered unstable. Compositionally-stable stands were separated into community types using COMPAH, an agglomerative clustering program (Boesch 1977).

RESULTS

The Chestnut Oak Community

Chestnut oak is the single major dominant (IV = 61.6; Table 1) in the chestnut oak community although a moderate number of tree species (14) were recorded. Except for white oak (IV = 10.1), other oak species had low importance values (< 7.7). Sourwood (Oxydendrum arboreun) was consistently present but at low importance values; this species was found only in chestnut oak dominated stands, and like chestnut oak is at the western terminus of its range. Post oak had an IV of only 6.9. Percent siminarity between overstory trees and understory seedlings and saplings was a high (76.8%) but five stands have values of 85 to 96 %. The chestnut oak were often moderately large (DBH = 35-45 cm) which resulted in a basal area of 21 m2/ha). Increment cores indicated that age of some of the largest trees exceeded 150 years.

Representatives of this community type were found on relatively dry (ave AWC = 8.2 cm H2O/profile), highly rocky (34 % rock) Cretaceous gravels of the Coastal Plain (Table 2). On steep (>25%). unstable slopes, the understory (seedlings, saplings, shrubs, herbs) generally had a low density probably from the excessive movement of surface rocks; here, sparkleberrry (Vaccinum arboreum) were the primary component of the mid-canopy. The soil often lacked leaf cover which exposed a surface of white gravel or patches of conglomerate. Moss and lichens were common. On ridgetops and more gentle stable slopes, a dense understory of blueberry (Vaccinum vacilans, Vaccinum staminium) often was present, the herb stratum had a higher species richness, and a thin leaf layer was present.

The Post Oak Community

In the post oak community, post oak had an IV of slightly over 50 (Table 1) followed by white oak and blackjack oak with an importance of 11.8 and 8.6, respectively. Chestnut oak had an importance value of only 3.8. Post oak stands consistently had a mid-canopy of sparkleberry and a few serviceberry (Amelanchier arborea). Percent similarity averaged 60.6; all stands except two had values above 50%; the highest similarity value was 76 %. Although the base area was moderately high (19 m2/ha), there were various canopy caps.

This community was restricted to loessal soil (Lax) of gently sloping southwest facing sites and occasionally ridgetops, and to relatively high elevations and slope positions (Table 2). A fragipan reduced water holding capacity for the profile to 10.3 cm.

DISCUSSION

We found that the compositionally stable stands of the chestnut oak occur on sites that are environmentally distinct from those of post oak. Both soil types (gravel vs loess with a fragipan) hold a limited amount of water (8-10 cm) for plant growth; thus, it is unlikely that white oak or more mesophytic species such as sugar maple or beech will invade these stands. Chestnut oak tolerates the low water supply and the twisting and distortion that occur in roots growing through gravel. Post oak also tolerates low soil water but unlike white and black oak or sugar maple, the roots are able to penetrate the fragipan and tap water stored in deeper layers.

Where the lax soil occurs on ridgetops. stands of mixed chestnut oak and post oak do occur. However, these ridgetop communities did not meet the selection criteria due to moderate disturbance and were not sampled in this study. We are presently gathering additional community data for this site type.

In a recent article, Fralish (1991) suggested that of the various forests that exist in southern Illinois, the stands on extremely dry sites appear to be the best representation of the pre-European settlement community. A comparison of present stands with data generated from General Land Office survey records indicates that although density may have increased in 200 years, the overstory composition remained the same. This concept also appears to apply to the chestnut and post oak communities of LBL. Although probably logged for wood for the iron charcoal industry in the mid 1800s and later, they have been relatively ignored for several reasons. The slopes of most chestnut oak stands are too steep for any type of farming activity and are a hazard to domestic livestock. Unlike chestnut oak which dominates forest over large areas, post oak occurs as isolated, overlooked pockets within a nearly completely forested landscape. Because of the slow tree growth and recovery from disturbance, neither community has much to offer for timber production.

We have referred to the chestnut oak and post oak communities as "woodlands" rather than savannas or barrens. In years past, the chestnut oak community probably had a much reduced canopy cover and was called a barrens. Because of low fuel loading most of the year, it is doubtful that fire could be successfully used to open the overstory canopy of chestnut oak woodland. At the moment, the stands on the Cretaceous gravel of the Mississippi Embayment represents a most rare and unique community. Moreover, with many of the trees approaching 150 years, we have recommended that they also be considered part of the old-growth forest presently being identified for preservation as part of the Biosphere Reserve.

The canopy of most post oak stands has a few canopy gaps but the thin crowns of post oak, blackjack, southern red and scarlet oak permit considerable light penetration. During presettlement time, our study stands probably had a more savanna appearance, On some sites where post oak IV is above 75% or where white and black oak are important components, a closed forest canopy and heavy litter layer hinder the growth of herbaceous flora. Typical open forest and barrens herbaceous plants are scarce in many of these communities, but common along LBL roadsides. Preliminary results of prescribed burning indicate that herbaceous flora richness could double under this disturbance. The seedbank, and roadside sources, may be enough to restore a more typical herbaceous understory to these sites.

LITERATURE CITED

Boesch, D. F. 1977. Application and numerical classification in ecological investigations of water pollution. EPA-600/3-77-033. Environmental Research Laboratory, U.S. Environmental Protection Agency, Corvallis, Oregon.

Bryant, W. S. and W. H. Martin. 1988. Vegetation of the Jackson Purchase of Kentucky based on the 1820 General Land Office survey records. P 264-276. In: D. H. Snyder (ed.). Proceedings First Annual Symposium on the Natural History of Lower Tennessee and Cumberland River Valleys. Austin Peay State University, Clarksville, Tennessee.

DeSelm, H. R. 1988. The barrens of the Western Highland Rim of Tennessee. P. 199-219. In: D. H. Snyder (ed.). Proceedings First Annual Symposium on the Natural History of Lower Tennessee and Cumberland River Valleys. Austin Peay State University, Clarksville, Tennessee.

Fralish, J. S. 1987. Forest stand basal area and its relationship to individual soil and topographic factors in the Shawnee Hills. Transactions of the. Illinois Academy of Science 80:183-194.

• F. B. Crooks, J. L. Chambers and F. M. Harty. 1991. Comparison of presettlement, second-growth and old-growth forest on six site types in the Illinois Shawnee Hills. American Midland Naturalist 125:294-309.
• S. B. Franklin, P. A. Robertson, S. M. Kettler and F. B. Crooks. 1993. An ordination of compositionally-stable and unstable forest communities at Land Between The Lakes, Kentucky and Tennessee. In: J. S. Fralish, R. P. McIntosh and O. L. Loucks (eds.). John T. Curtis: Fifty Years of Wisconsin Plant Ecology. Wisconsin Academy of Science, Arts and Letters. P. 247-267.

Francis, J. K and N. F. Loftus. 1977. Chemical and physical properties of the Cumberland Plateau and Highland Rim forest soils. U.S.D.A. Forest Service Research Paper SO-138:33-44.

Franklin, S. B. 1994. Late Pleistocene and Holocene vegetation history of Land Between The Lakes, Kentucky and Tennessee. Transactions of the Kentucky Academy of Science 55:6-19.

• P. A. Robertson, J. S. Fralish and S. M. Kettler. 1993. Overstory vegetation and successional trends of Land Between The Lakes, USA. Journal of Vegetation Science 4:509-520.

Harris, S. E., Jr. 1988. Summary review of geology of Land Between The Lakes, Kentucky and Tennessee. P. 199-219. In: D. H. Snyder (ed.). Proceedings First Annual Symposium on the Natural History of Lower Tennessee and Cumberland River Valleys. Austin Peay State University, Clarksville, Tennessee.

Ladd, D. 1991. Reexamination of the role of fire in Missouri oak woodlands. Pp. 67-80. In: J. E. Ebinger and G. S. Wilhelm (eds.) Proceedings Oak Woods Management Workshop. Eastern Illinois University, Charleston.

McCrain, G.R. and A.L. Grubb. 1987. An analysis of past and present vegetational patterns and historic parameters at Fort Donelson National Battlefield with recommendations for restoration and future management. Resource Management Co., Raleigh, North Carolina.

Table 1. Species mean basal area (m2/ha) and importance value (relative basal area) for eighteen chestnut oak stands and fourteen post oak stands at Land Between The Lakes, Kentucky and Tennessee. Species sequence is based on generally declining importance values.

Table 2. Environmental variable means for eighteen chestnut oak stands and fourteen post oak stands at Land Between The Lakes in Kentucky and Tennessee. Percent sand, silt, and clay data are from the A1 soil horizon.

	Community Type
Variable	Chestnut oak	Post Oak
Parent Material	Cretaceous Gravel	Loess
Soil Type	Saffell	Lax
Soil Texture	Clay Loam	Silty Clay Loam
Percent sand	27	19
Percent silt	44	52
Percent clay	30	31
Percent rock	34	18
Effective soil depth (cm)	52	48
Available water capacity (cm)	8.2	10.3
Distance to opposing slope (m)	377	425
Percent slope	21	14
Elevation	168	154
Aspect	All Aspects; Most Slope Positions	SW, S, W and Ridgetops

* Slope position was based on the percentage of between the ridge and nearest stream. In general: 1 = ridgetop, 2 = high slope, 3 = mid slope, 4 = low slope, 5 = drainage bottom or stream terrace.