Skip common site navigation and headers
United States Environmental Protection Agency
Great Lakes Ecosystems
Begin Hierarchical Links EPA Home > Great Lakes EcosystemsUpland Ecosystems > 1993 Oak Savanna Conferences > Wayne Lampa
Aquatic Ecosystems
EPA Region 5 Critical Ecosystems
Ecosystem Funding
Great Lakes Basin Ecosystem
Great Lakes Biological Diversity
Green Landscaping
Rivers and Streams
Upland Ecosystem



1993 Proceedings of the Midwest Oak Savanna Conferences1993 Proceedings of the Midwest Oak Savanna Conferences



Wayne Lampa  
(Formerly of Forest Preserve District of Du Page County)  

Gerould Wilhelm  
(Formerly of the Morton Arboretum)  

Conservation Design Forum, Inc.  
1250 E. Diehl Road, Suite 102 
Naperville, IL 60563 
Tel: (630) 955-0355  


Floristic changes were observed in remnant woodlands over the last decade. The study included 4 wooded tracts which were burned regularly, and one which was thinned and burned, contrasted with 4 which have remained unmanaged. In the burned tracts, the number of native taxa increased, the number of conservative taxa increased, groundcover vegetation increased, median light level increased, and adventive species decreased in relative importance. There were similar though slightly exaggerated changes in the tract that was thinned and burned. No significant changes occurred in the unmanaged tracts. Monitoring results indicate that larger tracts of woodland remnants would benefit from management. Recruitment of conservative taxa associated with high-quality savanna or forest systems appears limited to the immediate purlieus of the managed tracts or from within an existing seed or plant bank. It is possible that remnant unmanaged tracts are now so depauperate and their habitat so degraded, that they can no longer serve as viable corridors or conduits of conservative species. It is becommimg apparent that extensive management of contiguous remnant systems may be necessary if any individual remnant is to achieve its full potential for biodiversity and system stability. 


In presettlement times, timbered communities occupied only about 20% of Du Page County's area. These communities were composed primarily of oaks in relatively open stands where light was abundant enough to allow for a rich ground flora. Open oak forests often interfaced with denser, more mesic forests in protected areas and savannas or barrens on the prairie edges. 

Perhaps the first mention of Wisconsin oak openings was that by Charlevoix in 1761 who described the ". . . immense prairies, interspersed with small copses of wood, which seem to have been planted by hand" (Curtis, 1959). A little later Jonathan Carver reported in 1781 that "only a few groves of hickory and stunted oaks covered some . . . (of the prairies)" (Curtis, 1959). Keating, in 1824, spoke of "thin woods, which gradually disappeared," and were replaced by prairies (Curtis, 1959). 

James Fennimore Cooper's novel, The Oak Openings, written in 1848 from information supplied by notes made in 1812 at Kalamazoo, Michigan (Curtis, 1959): 

"The country was what is termed 'rolling', from some fancied resemblance to the surface of the ocean, when it is just undulating with a long 'groundswell'. Although wooded, it was not, as the American forest is wont to grow, with tall straight trees towering toward the light, but with intervals between the low oaks that were scattered profusely over the view, and with much of that air of negligence that one is apt to see in grounds, where art is made to assume the character nature. The trees, with very few exceptions, were what is called the 'burr-oak', a small variety of a very extensive genus; and the spaces between them, always irregular, and often of singular beauty, have obtained the name of 'openings', the two terms combined giving their appellation to this particular species of native forest, under the name of 'Oak Openings'." 

Fire played a major part in the development and maintenance of these ecosystems. For thousands of years, Native Americans maintained their landscape by means of annual burning. Hundreds of thousands of acres of prairie and forest were burned each fall as soon as the native grasses were dry enough to transmit the fire. The great conflagrations which swept the prairies did not stop at the woodlands. Indeed the Indians even burned the woodlands purposely. This served to keep them more open and shrub free. It also afforded ease of movement and attraction to game for hunting as well as some protection (easier to see the enemy coming). Natural fires did occur, but it was primarily Native Americans who maintained the character of the ecosystems until the time of settlement as Cooper noted. 

"These woods, so peculiar to certain districts of country, are not altogether without some variety, though possessing a general character of sameness. The trees were of very uniform size, being little taller than pear-trees, which they resemble a good deal in form; and having trunks that rarely attain two feet in diameter. The variety is produced by their distribution. In places they stand with a regularity resembling that of an orchard; then, again, they are more scattered and less formal, while wide breadths of the land are occasionally seen in which they stand in copses, with vacant spaces, that bear no small affinity to artificial lawns, being covered with verdure. The grasses are supposed to be owing to the fires lighted periodically by the Indians in order to clear their hunting-grounds". 

Not all of the early observers were agreed that fire was solely responsible. Owen 1852 thought wind action was also important, as shown in this passage (Curtis, 1959): 

"The country . . . is prairie, with oak openings. For about half the distance the growth of timber is very stunted; indeed, along the whole distance we saw no large trees. The dwarfish character of the timber does not appear to depend here upon the sterility of the land: it is true, the soil is siliceous, having been derived, in a measure, from the destruction of the sandstones . . . but it is not deficient in organic matter, and produces better crops than its appearance at first view would indicate. . . . Here I noticed that the most stunted trees grow in the more exposed situations. This seems to indicate that the strong winds which often sweep over the extensive prairies, exercise a considerable influence in retarding, and even suppressing, the growth of timber". 

John Muir gives a graphic account of these repeatedly burned oak plants (Curtis, 1959): 

"When an acorn or hickory nut had sent up its first season's sprout a few inches long, it was burned off in the autumn grass fires; but the root continued to hold on to life, formed a callus over the wound and sent up one or more shoots the next spring. Next autumn these new shoots were burned off, but the root and calloused head, about level with the surface of the ground, continued to grow and send up more shoots, and so on, almost every year until very old, probably for more than a century, while the tops, which would naturally have become tall broad-headed trees, were only mere sprouts seldom more than two years old". 

As recently as the late 1800's, there were eye witness accounts to the power of fire and its potential impacts upon the woodlands of the area (Curtis, 1959). 

"The testimony of Mr. J. E. Shaw, who has resided upon the prairies of Illinois for more than fifty years is equally to the point. He cites an instance on his own farm, where, forty years ago, when he took possession, there was a forest of large trees which was destroyed by fire, when a part of the burned district was again covered with trees, and a part was take possession of by the prairie grass, and in a comparatively short time could not be distinguished from the adjoining prairie". 

Presettlement plant communities were different from their contemporary counterparts. Our native wooded communities were as much controlled by the fires as were the prairies. These were very open in nature. The first settlers reported that the forests of the midwest were open enough that a wagon could be driven through them unimpeded (Pyne, 1982 ). 

"So dense was the original forest, it was claimed, a squirrel might travel from the Atlantic to the Mississippi from tree limb to tree limb without ever touching the ground. Cleared of this nineteenth century romanticism, the original accounts tell a different story. So open were the woods, one author advised with a touch of hyperbole, it was possible to drive a stagecoach from the eastern seaboard to St Louis without the benefit of a cleared road." 

It has been evident to us for many years that these communities were heavily overstocked with trees (primarily oaks) which had sprung up at the time of settlement when the fires were stopped. The resulting monoculture combined with heavy grazing, fire suppression, introduced species and other environmental factors to produce a decline in the original tree species and a depauperizing of the native ground cover. 

In spite of the fact that the oak and hickory canopy trees remained, few were to be found in the understory, which instead contained dense flushes of cherry, ash, elm and hard maple. It was obvious that oak replacement would be by these other types and that oak woodlands would disappear from our area unless some measures were taken to rectify the situation. Until now, however, we have not been able to document the nature and rate of the changes which have been taking place. 

We also noted that the woodlands were generally depauperate of wildflowers. We assumed this to be due primarily to past disturbance and the dense shade produced by the numerous small trees and shrubs in the understory. This has since been proven to be the case. 

In short the changes were evident. The true nature, magnitude and direction of change was not comprehended. We now know more about what is happening in the woodlands of the county. Unfortunately, the information shows that an overall decline is still taking place and that it is happening at a rapid rate. 


In 1979 a program to study woodlands was initiated with the placement of one-acre plots in a number of Du Page County's woodlands. Additional plots were installed in 1985 to monitor impending management in several more woodlands and savannas. During that same year, the study which originally included only trees and the overall plot flora, was altered to gain more information about the ground flora through the use of randomly placed quadrats. The methods used in the conduct of this studies is outlined below. 

For this study, eight wooded tracts were selected in seven Du Page County Forest Preserves. These particular sites reflect a cross section of woodlands in the county. Four plots were selected because they have been treated with fire between 1987 and 1991. Two to three prescribed burns were used in each of these plots during that time. The exact years and times of year differed from plot to plot because of the weather conditions. Four of the plots remained unmanaged during that time. None of these plots had been managed in any way prior to the study. Information is also included on a single plot which had been cleared of most of its understory in 1985, as an effort to reestablish savanna conditions under oaks and hickories which ranged from 190 to 210 years old. This plot was also burned three times between 1987 and 1991. 

Each plot was a 210 ft. square (approximately one acre), with as close to cardinal compass points orientation as possible. The plots were staked at the corners and the boundaries flagged with surveyor's tape. Everything found growing within the marked square was included in the overall survey. 

Random quadrat surveys of the flora were conducted in the plots in 1987, 1989 and 1991. This survey was conducted by locating the center of the plot and running four separate blocks of twelve random quadrats from that point; each of the points being a random distance and direction from the last placement. 

At each point the quadrat (0.25M2) was placed on the ground and the plant species present and approximate area covered by each were recorded. The plant species for each quadrat were then analyzed using the rating scale as expressed in Plants of the Chicago Region (Swink and Wilhelm 1979) to obtain quantitative data on the floral quality. A Braun-Blanquet (1932) cover/abundance number for the species present was obtained by using a 1 to 5 scale where area coverage was as follows: 

    • 1 = Occupying 1 quadrant and numbering 1 - few stems
    • 2 = Occupying 1 - 3 quadrants and numbering several stems
    • 3 = Occupying 2 - 4 quadrants with notable cover in each
    • 4 = Occupying 3 - 4 quadrants with regular density throughout
    • 5 = Occupying 4 quadrants and dominating all quadrants

From the plot surveys and subsequent transect data, it was possible to determine the following: 

  1. The average native plant quality of the total plot flora and the average quadrat flora. This is the average of the rating numbers assigned to each plant in the Chicago Region by Swink and Wilhelm (1979) 
  2. The average plant quality with the adventives (non-indigenous plants) included. 
  3. The native plant index ratings for the total plot flora and the average quadrat flora. This is the plant community index rating based on the average plant qualities as outlined in Swink and Wilhelm. 
  4. The plant index rating with the adventives included. 
  5. The number of native species noted during all of the flora surveys. 
  6. The total number of plants noted during the surveys. 
  7. The number and percent of total plants, native and adventive, occurring in the plots and in the quadrats as trees, shrubs, vines, ferns, forbs, grasses and sedges. 
  8. The percent of plants noted during the floral surveys which occurred in each of the various Coefficient of Conservatism classes. 
  9. Similarity indices which compare the flora lists for the plots to each other and individual plots over a period of years. 
  10. Relative Importance Values (RIV), which is a measure of the cover and abundance for each of the species noted during the plot surveys.

The available light reaching plants at the forest floor was surveyed in the same manner as the flora, but, at each random point a standard photographer's light meter was used to determine the amount of light (in foot candles) that reached that spot. Upon reaching the spot, the observer crouched (facing north to reduce reflected light from clothing) and extended the light meter beyond the body shadow at a height of approximately 18 inches and took the reading. Since average values are skewed greatly by one or two high readings, the best estimates of light values for the plots are derived from the median value which is then expressed as a percentage of the total possible visible light. 


The following observations were made in the four burned plots between 1987 and 1991. 

    * The number of native taxa per unit area increased by 26%. 
    * The number of native taxa encountered for the entire 48 quadrats increased by 37%. 
    * The number of native taxa with Coefficients of Conservatism of 4 or higher increased by 38%. 
    * The Mean Coefficient of Conservatism remained unchanged. 
    * The relative Importance Values of the Graminoids increased by 118%. 
    * The median percent of available light per unit area increased by 261%.

The following observations were made in the single plot which was both cleared and burned between 1987 and 1991. Note that this plot was cleared prior to initial survey in 1987. 

    * The number of native taxa per unit area increased by 12%. 
    * The number of native taxa encountered increased by 33%. 
    * The number of native taxa with Coefficients of Conservatism of 4 or higher increased by 43%. 
    * The Mean Coefficient of Conservatism increased by 17%. 
    * The relative Importance Values of the Graminoids increased by 185%. 
    * The median percent of available light per unit area increased by 29%.

The following observations were made in the four unmanaged areas between 1987 and 1991. 

    * The number of native taxa per unit area decreased by 36%. 
    * The number of native taxa encountered decreased by 2%. 
    * The number of native taxa with Coefficients of Conservatism of 4 or higher decreased by 14%. 
    * The Mean Coefficient of Conservatism decreased by 42%. 
    * The relative Importance Values of the Graminoids remained unchanged. Actually they were virtually absent. 
    * The median percent of available light per unit area increased by 67%. 


  • Fire, no matter how small or seemingly insignificant, stops the degradation of our savannas and woodlands. 
  • Burning opens up the understory by removing small trees and shrubs. This allows for more light and air to reach the ground flora. 
  • Opening of the understory and reducing the amount of shade allows for the return of oak woodland and savanna flora. Our studies indicate that a threshold is reached at approximately 6-7% of the available light. (Unmanaged woods generally have light levels of 1% to 2% of available.) 
  • Fire alone will probably not fully restore savanna conditions. Trees greater than four inches in diameter have not yet been affected by the fires. Mechanical or chemical removal of trees less than ten inches in diameter appears to be the only way to effectively restore woodland and savanna conditions. 
  • Many savanna species will return on their own (probably from the seedbank) to areas which are being managed. The most prominent among these are the sedges and grasses. 
  • Management by fire increases the number of native taxa per unit area while no management is often accompanied by a decline in flora. 
  • Management by fire will be accompanied by an increase in the number of conservative species per unit area. This is limited by the number which are present as plants or in the seed bank. 
  • The relative importance of graminoids in the managed areas will greatly increase. It is presumed that this is caused by the increase in light which also accompanies fire management. 
  • All of the above parameters are greatly enhanced when the burn management is supplemented by thinning of shrubs and undesirable trees in the understory. 
  • Adventive plant numbers and cover will increase with management. This is likely a resullt of the initial disturbing effect of the management and appears to be a temporary situation. 
  • An increase in the amount of light has been noted in several of the nonmanaged plots. This results primarily from the episodil loss of small elms to Dutch Elm disease. 


Braun-Blanquet, J. 1951, Pflanzensoziologic: Grunidzige der Vegetationskunde, Springer, Vienna. 

Curtis, J. T. 1959, The Vegetation of Wisconsin: an Ordination of Plant Communities, The University of Wisconsin Press, Madison, Wisconsin. 

Swink, F. and Wilhelm, G. 1979 Plants of the Chicago Region, The Morton Arboretum, Lisle, llinois. 

Pyne. S. J. 1982 Fire in America: A cultural history of wildland and rural fire, Princeton, N.J. , Princeton University Press. 


Begin Site Footer

EPA Home | Privacy and Security Notice | Contact Us