Much research in succession has focused on the journey of secondary succession from field to forest. For obvious reasons, researching phenomena that might be occurring over centuries rather than decades has been more difficult. But now that more than three centuries of intensive forest modification have passed, the longer-term consequences — such as the loss of the ground layer and plant and animal species — are becoming more apparent. Even more profound changes are occurring over even longer timeframes.
Many questions arise from this line of thinking, such as: Will the precise set of conditions that are needed to regenerate a majestic white cedar or tamarack swamp occur while their propagules persist? Will deer convert once-forested landscapes to brushland? Will the consequences of changing fire patterns further diminish the prevalence of oaks? Oaks have disappeared from many areas they once dominated, replaced by maple and beech where fire is fully suppressed. In the coastal-plain pine barrens landscapes, however, where fire was often too frequent for oaks, oaks are benefiting, at the expense of pine, from fire suppression.
For these reasons many in the restoration field view the reestablishment of historic successional patterns as the primary prerequisite for sustaining fragile landscapes. Indeed, restoration is sometimes referred to as “directed succession.” For some perspective on using succession as the context for management and managing succession as a primary strategy, let’s look at its principles and processes in more detail.
Frederic Clements (1916) is usually referred to as the father of succession theory, and his description, though outdated, still prevails as the popular perception of succession. He described succession as occurring in sequences, called “seres,” of discrete phases, each creating conditions necessary for the succeeding stage of the sere. In his view, succession is a kind of ordered unfolding of the potential landscape based on climate. He made little reference, for example, to the influence of the management by indigenous peoples or the impacts of current human uses.
Frank Egler (1954), also a successional theorist, recognized that the patterns of community succession are influenced by the presence of plants that have a head start because their seeds and other propagules are already in place and develop more quickly than those that must travel to that place. His theories of initial floristic composition and relay floristics held that the determining factor driving the unique patterns of succession in each place was the serial expression of those species that were established at the very outset because they determine much of the site context from that point on — by casting shade, by preempting space that would otherwise support other plants.
These patterns are easily observed in nature and have shifted with the growing prevalence of exotics. A meadow abandoned in central New Jersey today shows very different successional patterns than meadows in the same vicinity in 1940. Multiflora rose, exotic honeysuckles, and a variety of exotic invaders unfortunately make meadow management today more difficult than in the past.
Any change in the structure of the landscape will influence to some degree the future patterns of the landscape. Something as simple as planting a pole in the middle of a field will have consequences that will shape the future structure of the landscape. By providing a perch, a pole attracts birds and the plant propagules they bring with them. Succession is given a jump-start in such a place. Decades later, the largest and oldest trees in the forest will be found there, visible evidence of the early development of woody species in that spot.
Less well understood is the mechanism of “allelopathy,” chemical inhibition between species and individuals, in successional change. Virtually all plants exude chemicals from their roots, in their decaying leaves, and in the moisture conveyed across their leaves. These substances may affect other plants. The difficulty of growing other plants with black walnut and in the barren zone within the dripline of many eucalypts is legendary, but allelopathic processes also affect forest succession as well as the compatibility of agricultural crops, determining which plants can be grown together. Allelopathy influences which species can become established and persist in any given landscape. Perennial herbaceous plants such as goldenrod and aster actually inhibit the germination of tulip poplar and pine, thereby prolonging the open phase of a meadow and postponing their eventual demise under forest cover (Brown and Roti 1963). The Norway maple has been observed to be strongly allelopathic. Its fallen leaves can retard growth of many species, so that the ground beneath its thick canopy is often markedly bare. The tree-of-heaven may be even more potent. Extracts from its leaves and twigs have been shown experimentally to act as an effective herbicide; commercial applications are currently being evaluated (Heisey 1996).
Jack McCormick, one of the most important plant ecologists in the field of allelopathy, noted in 1968:
[S] ufficient evidence now is available to suggest that allelopathy may be of great significance in succession or lack of succession. In addition, autotoxicity may . . . be involved. Autotoxicity may result in inhibition of seedlings in the vicinity of mature plants of the parent species and, thus, may play a major role in population density control. It may also produce reduction of stature of annual plants growing on a site formerly occupied by dense stands of the same species, fairy ring configurations of perennial clones, and other effects (32–33).
Managing allelopathy may offer intriguing opportunities to influence succession but will require many field trials. For example, mowing and mulching of perennial grasses with their own freshly cut stalks may stunt their growth and be a component of an effective control strategy for some grasses such as common reed and miscanthus. Cutting and removing the end-of-season vegetative growth of goldenrod and aster may reduce their ability to retard the germination of woody species. Removing the herbaceous cover from a mown meadow may accelerate the return of woody species.
All management is an effort to direct the process of natural change, to arrest it by mowing and applying herbicides or to accelerate it by planting and irrigating. Here in the East, where rainfall is more abundant and evenly distributed seasonally than in the great open plains in the central portions of the United States, almost all grasslands, whether they contain native prairie species or exotics, require management to keep trees from developing and overwhelming the herbaceous species, eventually returning the landscape to forest.
Human uses more frequently arrest the speed of succession than accelerate or promote it; consequently, the landscapes around us are getting younger at the same time that they become more fragmented. Cropland is held forever in a juvenile state, sustained primarily by resown annual species. Wildflower meadows need to be mown or weeded of woody volunteers at least periodically or a forest will grow. Lawn is perhaps the most extreme example of this pattern. In a landscape that has the potential to support a great and multilayered forest it takes a lot of labor, energy, fertilizer, and pesticide to prevent that forest from growing. The heavy watering and chemical use that lawn requires in turn means excessive runoff, much of which is highly enriched or contaminated, with adverse impacts on remnant natural areas. The resulting landscape performs none of the functions of forest and supports almost none of its creatures.
All the landscapes around us represent some moment in time on the successional continuum, whether we are performing conventional or ecological management, and it is from the perspective of that particular moment that we want to manage each landscape. The stage of succession will determine where the landscape is trending. An important and simple distinction is between those landscapes that are still primarily herbaceous and those that are predominantly woody.
The most common activities directed toward managing succession in an effort to control the direction of change include adding species or taking them away as well as manipulating environmental conditions by burning or flooding by impoundment. In the case of forest restoration projects currently under way in Central Park, all of these actions are taken simultaneously. Propagules of all native species mentioned in the historic record are added to the site because there are no longer any natural sources of these propagules. At the same time, exotics are removed to reduce competition with natives.
In order to manage with succession, you will often find it useful to identify the seres, or sequences of successional stages, that occur on your site as well as determine as clearly as you can what your successional goals are. For example, are you trying to arrest succession at a meadow stage or revert to a grassland from a vine scrubland? Are you trying to accelerate succession to create conditions more favorable to species of the primary forest?
Although the phases associated with succession are more gradational than discrete, simplifying your management review by generally summarizing the characteristic stages in your landscape can be helpful. The following description is simply a guideline for what is typical of many temperate second-growth woodlands. In your own region, you should describe its special and characteristic successional changes, specific plants, and plant–soil associations. In most environments there are numerous gradients — for example, the gradients between moist and dry soils or between highly disturbed and more natural sites.
Herbaceous landscapes are typically small in scale, ranging from short turf underfoot to knee-high or even shoulder-high and taller grasses and wildflowers. When herbaceous landscapes are effectively stabilized, you cannot see the ground because the herbaceous cover is so dense. Natural herbaceous landscapes typically occur after a disturbance such as fire or vegetation clearance and are relatively temporary. They may persist where the land is too wet or too rocky to support trees. Modified herbaceous landscapes such as flower gardens and lawns require maintenance to arrest natural succession, or they too will succeed to woody species.
We have many opportunities to enrich the range of herbaceous landscapes around us by taking a cue from natural succession. By confining lawn to the smallest area necessary, we can then explore alternatives to turf such as a “greensward,” a taller grass lawn in which grasses and short wildflowers are intermixed, which was typical in the last century. Tall-grass and wildflower meadows, prairies as they are also called, are another possibility. In addition to reducing the amount of lawn, we can restore many waste places and weedy landscapes that are merely the result of poor maintenance but that with management could become quite rich. Today there is renewed interest in meadows, both as scenery and as a cost-effective alternative to turf. Maintaining a meadow costs on average only $50 per acre a year, less than one-tenth the cost of lawn. But we have lost the historic art of meadow management in our recent attachment to lawn and must relearn the techniques for tending meadows.
Along the continuum from grassland to forest, there is a long period of transition as the maturing woody plants gradually assume predominance over the herbaceous vegetation. Woody plants, of course, have been there since the outset, less visible amid the tall grasses and wildflowers, but they take on a dominant character as they mature. Such landscapes are known as “woody oldfields” or “savannas.”
John Curtis of the University of Wisconsin Arboretum, where he initiated the first prairie restoration, defined “savanna” as grassland with up to 50 percent forest canopy (Curtis 1959). In the Midwest the savanna landscape occurs as a transition between the limits of the eastern forest’s range and the drier grasslands of the western prairies and the plains. Indigenous peoples historically burned the midwestern savanna in the fall, creating and sustaining the mix of grassland and woodland of the savanna landscape type.
In the East, the term “savanna” is less commonly used, but a savanna-like condition of about half trees and half herbaceous species frequently occurs for a period of time on the successional continuum. Another name for this landscape is “woody oldfield” (literally “old field,” that is, abandoned pasture or cropland). In an oldfield abandoned from agriculture, for example, this transitional semiwoody condition might last up to several decades.
At the outset a woody oldfield is dotted with small trees in an herbaceous matrix, but soon the growing size and number of trees creates a closed canopy in patches at first and then more generally. Although historically less extensive than those of the Midwest, savanna or woody oldfield landscapes in the Eastern forest support species that are adapted to this temporal niche, including many now-rare species as well as the once far less abundant white-tailed deer. With management, you can sustain such a landscape indefinitely, providing an alternative to turf and specimen trees where an open parkland character is desired.
Woodlands and forests are landscapes that are predominantly trees and other woody plants where the canopy cover ranges from greater than 50 percent to completely closed. They have a structure composed of several layers of woody plants, from canopy and understory trees to shrubs and a ground layer. Herbaceous cover is typically relatively limited. In a forest you can often see the ground, which is usually covered by a layer of leaf litter. If, however, the soil is bare, that is usually an indication of disturbance.
The term “forest” usually refers to multilayered, multiaged, and multi-specied wooded landscape; “woodland” generally is more open and younger in age and structure than a forest. Forests and woodlands are stabilized by the multilayered system of roots of multilayered vegetation as well as a multilayered ground topped with a litter of leaves and other forest debris. The ground-layer vegetation may be patchy or only seasonally apparent. Soil stability does not depend on very dense vegetation, stem to stem, as in an herbaceous landscape, but rather on overlapping systems of stories, canopies, and a complex root structure linked with ectomycorrhizal fungi.
Restoration of woodlands and forests may entail the addition or removal of either selected layers of the woodland or specific species, as well as changes to environmental conditions, to accelerate or retard the processes of natural succession. The management team may focus on reestablishing a more continuous canopy where forest cover is too spotty or on enhancing the diversity of patches depending on the current condition of the landscape and their overall goals.
Within the forest matrix are a variety of patches that reflect different environmental conditions such as soil chemistry and available moisture as well as past history such as blowdowns or active browsing. Through management, you can mimic a variety of different environmental conditions to favor selected species or meet specific functional requirements of the landscape. Forest glade, forest edge, and light forest are special management options that may be suitable to places on a site.
The term “glade” is often used to describe an area within a forest where vegetation is less dense and thereby provides an opportunity for a special plant species to thrive and creates a place of pause for the walker. The conditions favoring a glade may occur naturally in the landscape for a variety of reasons. The light forest or glade character often results when a layer of the forest landscape is missing — for example, where a large canopy gap has fostered a grove of locusts growing in a ring or where the largest and oldest tree in the forest has limited the other vegetation beneath its sway to an ancient carpet of woodland ephemerals. These places draw the eye, and hence the walker, to stop for a moment and observe the interplay of light and shade, texture and pattern.
Public concern for personal security is an issue in many woodlands where public use is high, such as in urban areas, but you can increase visibility in the landscape in selected areas without resorting to wholesale clearance of all low-level vegetation. Limited reduction of vegetation to create a light forest condition is one way. Management for this purpose consists of a careful editing of selected plants in the forest layers to enhance a sense of an opening within the enclosed setting. By keeping a managed glade small in area, rarely more than the width of a single tree canopy, you can retain the sense of forest and canopy closure above. Or you can extend light forest over a slightly larger area, such as the margins of a path or at the juncture of two or more paths, without creating a large gap. Forest edges are generally the most managed places in the landscape because so many sources of disturbance stem from them. A good rule of thumb is to “seal the edges” by developing a dense, multilayered edge to minimize the distance that edge impacts extend into the forest. This strategy is especially important where new edges have been created and where control of exotics is necessary. In many other cases, the goal will be to accelerate succession to re-create conditions that are more similar to the ancient forest of dead wood, pitted terrain, and deep humus.
Keeping a perspective on change is vital in any management program. Each moment in time on the successional continuum presents different opportunities and constraints. Management affects the range of successional landscapes in each place and the potential of future landscapes.
Roadside landscape maintenance exemplifies how management can be completely in conflict with the processes of succession in native communities. For instance, wherever a roadside is not managed as turf, the usual practice is to periodically clear away the woody vegetation. This habit of intermittent brush clearance along roadsides guarantees that the development of woody species cannot proceed naturally At the same time, the landscape is mown too infrequently to sustain grasses and wildflowers. Weedy species tolerant of frequent perturbation, such as Japanese knotweed, are the inevitable result.
A key difference between maintenance and management is that maintenance, like housework, must be repeated endlessly. Mowing, for instance, is continually needed: we don’t expect a lawn to mature and then need less mowing. However, we can often expect management to diminish or at least shift focus as the restored landscape becomes more established and as disturbance is better controlled. A wildflower meadow may always require some mowing or burning or other woody control, but with time our efforts may be directed more to species enhancement than to elimination of exotics such as multiflora rose. In Central Park, after only a few years of invasives management, workers were able to shift focus from removing Norway and sycamore maple to restoring microorganisms necessary to soil quality.
In conventional management, the assumption is that if you are diligent at maintenance, you can be sure of the results. We can expect a green golf-course-type lawn if we mimic the management practices recommended for such turf. The assumptions for restoration management are somewhat different, even though restoration does involve a lot of repetitious activity The major distinction is that a restoration program should change continuously in response to changing site conditions. The more rote the activities, the more you should look for negative impacts that have been taken for granted, such as overriding the seasonal and longer-term natural variations that are necessary for the recruitment of highly specialized species.
Frequent changes in management direction can be stressful to a landscape and almost always favor opportunistic species. The more frequent the disturbance, the more simplified living systems are likely to become. Deciding to release a meadow — for example, letting it go to woody species such as multiflora rose without any follow-up management — will achieve a more degraded meadow. Similarly, clearing exotics in a forest and failing to restabilize, replant, and monitor as necessary may only invite further invasion in the new openings. We need to learn to examine any proposed changes in management direction closely and be able to defend them based upon monitoring. It is axiomatic that it is easier to sustain existing native communities than it is to reestablish them. It is also easier to achieve a minor shift in the composition of a native community than it is to go from an exotic-dominated to a largely native landscape.
More active and timely intervention is a goal of restoration, but also keep in mind that it is easy to overreact. Removing exotics too aggressively can be as damaging to relic native populations as it is to the target invasives. Planting more than is necessary for restabilization may overwhelm natural seeding and can obscure natural patterns of reproduction. Restoration is a form of research that requires an openness to the new and previously unseen. Observation of interrelationships and influences in the landscape is the manager’s most important skill. All too often we ignore the mechanisms that are driving change in the landscape rather than focusing intently upon them.
All forests have a fire history, whether the cycles are long or short, catastrophic or light. Land-use and settlement patterns have dramatically altered successional fire-related patterns that persisted for millennia. For example, infrequent fires in forest expanses sometimes created large gaps that provided opportunities for species like white cedar. These conditions necessary for regeneration do not occur in fragments. Some areas, such as sandy coastal landscapes with mixed oak and pine, often burned more frequently in the past. Today, more mesic woodlands typically burn either more frequently or not at all. Urban grasslands may burn yearly or even more often, while more rural ones may never burn but are regularly mowed instead. Without these historic patterns of fire, succession has been dramatically altered.
Attitudes toward fire illustrate how management perspectives are shifting in the direction of restoration. We now recognize that fire suppression alters plant succession and results in far more devastating conflagrations over time because excessive amounts of accumulated fuels build up. Practices that foster more natural fire and hydrologic cycles also make natural habitats more resistant to invasion by some diseases and pests, both introduced and naturally occurring. The restoration of native shrub and understory layers and the reestablishment of a natural fire regime appear to be more effective in controlling pine bark beetle, for example, than vigorous eradication and clearance efforts. Maintenance of natural fire cycles in the Pine Barrens of New Jersey is required to sustain the fire-dependent or fire-tolerant species in this landscape.
In addition to providing beneficial effects that are difficult to replicate in other ways, fire is a relatively low cost management tool. Meadow managers generally acknowledge that fire is the fastest route to a more diverse wildflower meadow or a denser stand of native grasses. Management approaches range from virtually natural wildfire cycles on large reserves to highly controlled burns. Even on small sites we can try to replicate processes that without management would not occur except in very large reserves. Ultimately, the goal of all fire management in restoration should be to restore as close to a natural fire cycle as possible given the landscape context of the site, whether it is rural, natural, or bordering a residential area.
In larger forested tracts, the reestablishment of more natural fire regimes should be given high priority. A pulsed cycle, with varying intervals of burn that closely mimic natural fire patterns, rather than regular intervals, is preferable. Even on smaller sites, management can seek to reestablish a more natural and varied pattern.
At the Connecticut College Arboretum, William Niering, an ecologist and former director of the arboretum, conducted long-term studies on prescribed burning in both woodlands and grasslands. In one forest tract, annual burning was undertaken for the first few years to reduce fuel accumulation. That also served to renew many herbaceous species and resulted in increased germination of oaks and other species with previously poor recruitment. At this point, Niering withheld fire to allow the new saplings to develop to a sufficient size that they would not be killed by a light ground fire. Spotted wintergreen and other woodland ephemerals spread notably across the newly opened ground.
In most built-up areas, light controlled burns limited to restricted sites will be all that is feasible, but you can still mimic a natural fire. Where smoke must be severely curtailed, burning in very small scale patches can help to create greater variability in landscape patterns and to replicate the conditions of greater fire intensity.
The restorationist should consider using fire more extensively as a management tool for many tasks that are now accomplished by more destructive methods, while maintaining stable cover and reducing long-term management costs. For example, where visibility in woodland areas is currently managed by mowing, consider using fire as an alternative to encourage herbaceous species and compact, low shrub growth while sustaining sight lines in limited areas. Native grasses, such as broomsedge, are long-lived perennials invigorated by periodic burning, which at the same time controls new tree growth that would suppress the grasses. Prescribed burning may also reduce the impacts of allelopathy because the aerial parts of the plants are consumed, rather than left behind as a mulch, which is usually the case with mowing. In turn, it may also extend the reign of the plants already established on a site by reducing the rate at which their litter accumulates and the autotoxicity resulting from allelopathy. In this way, fire also may limit the introduction of some exotics into the landscape.
Use of controlled burning entails a sometimes-lengthy permit process because of air-quality concerns. Local air management regulations vary and must be reviewed carefully. Of particular concern are possible hazards of smoke reducing visibility on roadways, for example. Requesting the assistance of local fire departments gradually develops a wider network of personnel trained in controlled-burning techniques who can help you ensure adequate safety and compliance with insurance requirements.
Adequate control of a prescribed burn is dependent on a system of firebreaks, which may include natural features, such as streams and wetlands, or built features, such as roadways and lawn areas. Where new firebreaks are required, careful review is mandatory to ensure that the firebreak does not serve as a route for disturbance, disrupt natural drainage, or otherwise adversely impact the landscape.
The Nature Conservancy has made a major commitment to the restoration of fire in the landscape and conducts training workshops on prescribed burning and maintains burning programs in many of its reserves. The National Park Service also recognizes the importance of fire as a management tool and has developed a manual on wildland fire management for the control of wildfires and the management of prescribed and research burns (National Fire Policy Review Team 1989). Major topics included in this manual are the identification of roles and responsibilities of governmental agencies, procedures for fire analysis, documentation, and staff training and distribution, as well as guidelines for wildfire control and management objectives for prescribed burning.
Controlled burning has optimal results when it reflects seasonality and historic patterns of wildfire as well as fire management practices by indigenous people, which in the past were complementary to and shaped wildfire patterns. As a practical matter, though, managers today schedule most prescribed burning for the winter and early spring for safety reasons, especially where fire frequency has been suppressed and fuel loads on the ground are excessive. In general, spring burning favors woody species that sprout. Fall burning tends to favor herbaceous plants, as well as nonsprouting species, like many conifers, that require a specialized seedbed. Fall fire is also more typical under natural regimens, but it may not be feasible because of fuel loads. Where the prolonged absence of fire has resulted in overgrowth, conservative burning practices are often necessary, at least initially, because of excessive fuel accumulations. The prolonged absence of fire in some areas affects the kinds of techniques that can be utilized. Burning downslope with firebreaks and backburns is often necessary until fuel loads no longer pose unnecessary risks to more aggressive burning techniques.
Burning upslope is swifter and more characteristic of wildfire. Headfires that move with the wind are cooler at the surface but hotter at about 18 inches and above and therefore more effective for killing shrubs and trees while burning deadwood and woody debris. Backfires — that is, fires that move against the wind — are slower and hotter at the surface and reduce damage to the overstory while consuming large fuel loads. Fires that burn sideways to the wind are safer than headfire but not as slow as backfire. Ultimately, the objective is to re-create historic cycles and scales of wildfire as well as burning like that of indigenous people. Much more research is needed to document historic people’s management of fire in the landscape as well as their other practices, since the habitats of the lands they once managed are those we seek to sustain and restore.
Initiating a fire management program is not easy. Agency personnel and government guidelines are often so cautious that a natural pattern of fire is difficult to achieve. Volunteers have often had to reinvent the techniques of fire management as they go along. Nonetheless, the effort is well worth pursuing.
A key component of the effort to sustain and restore the natural landscape is the monitoring of population trends in order to evaluate succession in that landscape. By monitoring population change you can also evaluate the success of your efforts to reintroduce or protect selected species or to foster specific communities. Even a decision to take minimal action, such as to allow for natural recovery after a disturbance, should be supported by ongoing monitoring. As McCormick (1968) noted, “The most valuable evidence from which to develop successional theory is obtained from direct long-term studies of vegetation on specific areas.” The forest of the future is growing today, and change will be expressed over decades and centuries.