Chapter VIII - Conclusions

In order to understand the place the presettlement Piedmont savanna has in the modern landscape, I have in this thesis considered visual characteristics, ecology, cultural and natural history and the relationship of human preferences to savanna landscapes. Summarizing these aspects points out the value, and the degree of applicability this landscape has two hundred years after its demise.

Evidence in the cultural environment around us suggests that people prefer savanna-like landscapes. This preference is manifest in landscape types such as suburban yards, parks, campuses, and orchards where vegetation structure reflects the scattered trees and open understory of the savanna. Appleton explains this phenomenon as being a result of our bioevolutionary heritage which gives us the affinity for landscapes high in prospect and refuge. Prospect and refuge were key elements in early human survival as our ancestors progressed from being reclusive tree dwellers to more aggressive ground dwellers in the savannas of Africa. The Kaplans, for similar reasons, propose that landscapes are most preferable when they make sense and offer fulfillment, or, more specifically, coherence, complexity, mystery, and legibility. When savanna landscapes are measured in these terms, their high ranking shows that they are indeed favored by humans.

Part of this work has focused on the ecological history of the Piedmont and people's perception of it. Because few people are aware that Piedmont savannas existed or that they played a role in the region's history, it has been important to show not only evidence of their occurrence, but also reasons for their demise. The existence of the Piedmont savannas has been well documented in the many historical citations from Bartram, Byrd, Lawson, Lederer, Hawkins and Spangenberg and the writings of later historians. Savannas may have come into being due to climatic conditions, but their spatial distribution over time has been modified greatly by humans. Piedmont and southeastern grasslands in general, whether a result primarily of Indian or natural burning, were closely linked with patterns of Indian culture; the migration and presence of presettlement fauna such as bison, elk, and deer; and the direction and rate of European settlement. The immediate disappearance of these landscapes upon European settlement was caused by fire suppression resulting from fire compartment fragmentation.

During the completion of this thesis an important article was published by Barden in Natural Areas Journal which further discusses the history of the Piedmont savanna.

Understanding presettlement Piedmont savanna ecology is essential, not only in explaining the existence of savanna communities, but also as a model for any related reconstruction of such an environment. Of all the factors that contributed to the maintenance of savanna landscapes, fire was without a doubt the most important. It is recognized that Indians may have been the major cause of fire, though in some cases fire from lightning may have been responsible for maintaining an open landscape. If one considers that by burning, humans were probably the most important agent in shaping the structure and composition of the presettlement landscape, then it might be appropriate to argue that any restoration of such a landscape would, in fact, be a demonstration of a cultural landscape, more so than of a "natural" one. In a way, this aspect of the presettlement condition brings the often diametrically opposed views of culture and nature closer together.

Where frequent burning was combined with the more difficult growing conditions found on basic soils high in montmorillonite (shrink-swell soils), the most open and often prairie-like examples of the Piedmont savanna occurred. This type of soil, though not the most common, can be found scattered throughout the Piedmont. In South Carolina, North Carolina, and into Virginia, roughly along the Interstate 85 corridor, these soils are most common, and reflect the probable former distribution of prairie-like landscapes in that region. It is probable that the other upland soils of the region, mainly Ultisols, supported woodland savannas rather than prairie. The exception to this would have been in the case of proximity to Indian settlement. Areas where high Indian populations existed, especially in lowlands where rich soils provided the best nutrients for crops, were sometimes very open, to the extent that they might have resembled today's farmland.

Species composition of the presettlement Piedmont also reflected the presence of fire. The oak-hickory-pine-type woodland of that time was dominated by species more fire tolerant than those dominating today's forests. In savannas, fire would have reduced or eliminated the understory. A grassy stratum would have been common to both the open woodlands and prairie-like areas of the presettlement landscape. This grassy ground layer would have reflected the composition and many of the species of the tall grass prairie.

The degree to which the landscape appeared open was a reflection of the influences of fire frequency, soil moisture, exposure, and soil structure. Depending on fire frequency, the visually dominant tree canopy layer may have varied from patches of low trees over a grassy ground plane in more xeric conditions to patches, peninsulas, or even distribution of taller trees over more mesic conditions. Distribution patterns in tree canopy and ground plane species would have reflected the often drifting patterns common in natural phenomena. Sharp transitions or edges would have occurred in relation to abrupt changes in soil, slope, and moisture conditions, or perhaps where intense Indian activities occurred. Both grassland and woodland would have intermingled and graded into cane brakes, a nearly ubiquitous fire-dependent feature of wet soils along drainages. All of these various gradations and edges abundant in the Piedmont savanna would have supported an especially rich species diversity. Savanna ecotones, or edge environments, though they do not support all grassland and woodland species, support a greater number and variety of creatures than can either the grassland or woodland community alone, by creating habitat for both feeding and nesting.

At times the presettlement Piedmont savanna may have been similar in appearance to the pine flatwoods of the Coastal Plain, while at other times to the oak savannas on the edges of the Midwestern prairies. Savanna landscapes would often have appeared as pleasant scenes of open, sometimes distant views, with accentuated lighting and atmospheric effects not to be seen in the denser forests. It is unfortunate that the landscape painters of the Hudson River School came along a hundred years too late and a few miles too far north to represent these scenes for us.

Designing or restoring Piedmont savanna landscapes presents an opportunity to demonstrate an aesthetically, ecologically and historically rich landscape heritage. It is important to provide a model or demonstration for such a restoration design, first because it allows the public to perceive the aesthetic value of an otherwise unseen landscape. Second, the public needs to understand that this landscape is part of our natural and cultural Piedmont history. Third, it provides an opportunity for the public to learn about the ecological functions of the savanna. Fourth, there is a need to study the Piedmont savanna to determine its long term value as an ecologically sound model for landscape design. Finally, creating these landscapes can provide habitat for some of the endangered species that were once dependent on open, fire-maintained landscapes.

It should be considered, however, that there are some complicating issues in savanna restoration. These include context, scale, and time. First, it is important to use the appropriate amount of "natural function" to maintain the savanna according to the context in which it is located. For instance, in a small garden-like savanna in a neighborhood, one might need to forgo burning and grazing in favor of mowing and manual weeding. If a large-scale landscape in a rural setting is the application site, then the unsightly effects of the first year's slash and burn may be more acceptable while the characteristic long-term aesthetic qualities are developing. On the other hand, when the intention is to demonstrate these ecological functions, the site should be in a more visible location. In this case, plenty of information must be provided to the public in order that they may develop an ecological aesthetic that will allow them to appreciate the temporary roughness of the savanna's appearance. This is especially important in the early years of the installation, when it may be at a visual low point.

The amount of time needed for a savanna installation to attain a mature appearance can add up. Fire-dependent communities, quick to disappear with fire cessation, are slow to redevelop once fire is restored. Some applications may be able to reach the appearance of a mature savanna sooner than others. For instance, where the appropriate tree canopy already exists, a herbaceous layer can be established in only a few years. If a tree canopy must be installed, then faster-growing pines could allow the savanna to reach a mature appearance in only a decade or two. In the demonstration savanna, viewers should be informed that the slow-growing savanna climax species such as post oak and black jack oak will take many decades to reach their full visual potential.

Perhaps in the end, considerations having been given to history, ecology, aesthetics, and applicability, the most rewarding aspect of demonstrating the Piedmont savanna might be found in the imagination of the viewer, who, as he or she gazes upon a ghost of a landscape long past, may catch a glimmer of salubrious prospects over undulating open woods where buffalo and wolves roamed and humans drank from the clear waters of Piedmont creeks.

back to top