Approaches to Ecologically Based Forest Management, PART I: Ecological concepts upon which forest management is based

Every forest is in a state of change at all times. This change is not always apparent through direct observation even to a trained forester. Most foresters simply do not have the opportunity to work in the same forest for more than a decade - too short a time, in most cases, to directly observe changes in composition and structure. Nevertheless, forest managers must understand the factors and conditions that control such changes if they are to achieve the desired management outcomes. Understanding natural forest dynamics in a particular region should be the foundation of every management action.

Silviculture, by definition, involves planned change. If we do not understand the natural dynamics, we cannot effect a planned change. Too often, foresters assume that an existing, desirable forest cover type can be maintained by the same silvicultural treatment that was successful in another area or on another site. This, of course, may not be so if the two stands exist on significantly different site types or have resulted from different disturbances. Examples of such cases are probably known to every forester. Thus, a forester should not begin with the unqualified assumption that maintenance of the current cover type should automatically be the future management objective. Often, conversion to some other composition or structure, by taking advantage of natural trends, may meet or exceed management expectations and simultaneously improve ecological conditions.

Although every geographic region is unique in terms of its forest composition and the role each species plays in forest dynamics, some general models of forest change can be discussed. When speaking of forest dynamics, we usually envision some model of shade tolerant species replacing a less tolerant one. However, this classic model of succession represents only one of many modes of forest change. Some of the better understood modes of change are described below.

Often, when a forest is severely disturbed by natural forces or by complete removal of trees through logging, the "new" forest is composed of different species compared to the "original" forest. These "pioneer," or early successional, species making up the "new" forest are generally (but not always) intolerant of shade and will not reproduce in the understory to form another generation. Only another disturbance or silvicultural intervention can re-initiate this stage. Without such a disturbance or intervention, other more shade-tolerant species gradually replace the pioneer forest. Depending on the region, this replacement may occur through two or more stages. Ecologists refer to this type of succession as "relay floristics" (Figure 1). The course of this type of succession is also strongly affected by site type and other factors to be discussed in subsequent sections.


	Figure 1. An example of "relay floristics" type of succession. In this case, balsam fir is replacing trembling aspen.



In many cases, forest composition does not change so completely or in a singular direction as it does in the case of the relay floristics model described above. Many tree species are moderately tolerant of understory conditions, particularly in juvenile stages, and persist in a stand for some time. These species are able to take advantage of openings in the canopy that occur through either small-scale disturbances or death and removal of both individual and small groups of trees. This mode of replacement is strongly dependent on the composition of the stand and site type.	Oak seedling taking advantage of a canopy gap.
For example, if a stand is a mixture of midtolerant (e.g., red oak - white ash) and tolerant (e.g., sugar maple) species, and occurs on a rich mesic site, any canopy gaps that occur through death of single trees are likely to be "captured" by tolerant species. This is because tolerant species are likely to be better represented in the reproduction layer and their growth rates are optimal on such sites. On the other hand, if a similar stand develops on a drier, less fertile site, the mid-tolerant species have a greater chance to fill gaps because on these sites their growth rates exceed those of the more moisture- and nutrient-demanding tolerant species. This situation is depicted in Figure 2. Stage 1 in Figure 2 represents a sugar maple-dominated stand with occasional white ash and red oak mixed in. The reproduction layer is likewise dominated by sugar maple, but a few white ash and red oak seedlings can be found. Stage 2 shows a canopy gap where a mature tree died. The seedlings that are taking advantage of this gap are those of faster-growing white ash and not the more ubiquitous, but slower-growing sugar maple. The gap was not large enough to also accommodate a red oak seedling at the edge of the gap. Stage 3 shows the original gap filled by two white ash trees and the formation of another gap. This time, no white ash or red oak seedlings were present in the location of the gap, and the space was filled by sugar maple saplings. The white ash and red oak seedlings still shown in the understory of stage 3 will not survive the suppression. In the absence of major disturbance in this type of stand, continuous gap replacement maintains a mixed composition and multi-aged structure.

The so-called climax, or late successional, forest communities are thought to be self-replacing due to the ability of canopy species to form advance regeneration. However, such advance regeneration is not continuously present in many types of climax communities, or if present, it is not continuously advancing into the canopy layer. Instead, such communities are transformed through a series of recognizable stages: stand initiation, stem exclusion, understory reinitiation and old, multi-aged community (Figure 3):

Figure 2. Canopy gap replacement as a function of chance occurrence of seedlings in the gap area and differential growth rate of species on a dry-mesic site.

SM=sugar maple
WA=white ash
RO=red oak

See text for further explanation.


Stage 1. Stand initiation. This follows major disturbances, such as catastrophic wind, fire or clearcutting. The open space becomes filled with individuals that arrive by seed (e.g., paper birch, yellow poplar, aspen, cherry), stump sprouts (e.g., oak after fire) and root sprouts (e.g., aspen after clearcutting), or that were present as advance regeneration (e.g., sugar maple or other shade-tolerant species after a tornado or logging removes the canopy).	Aspen root suckers.
The individuals are part of an age group called a cohort. This stage ends when the canopy becomes continuous and trees begin competing with each other for light and canopy space.

Figure 3. Changes in stand structure without species replacement. See text for explanation.


	Stage 2. Stem exclusion. During this stage, the canopy is dense enough to prevent new saplings from growing into the canopy - there is no space available for new canopy trees. The canopy continues to have only one dominant cohort, with a relatively smooth upper canopy surface. Competition among trees is intense and density-dependent self-thinning is the major cause of mortality.
Crowns are small enough so that when one tree dies, the other trees are able to fill the vacated space in the canopy by expanding their crowns. The duration of this stage varies with species and geographic region. For example, in the Lake States and the Northeast, this situation continues for 75-150 years in northern hardwoods and red or white pine stands, but may last only 20 to 40 years in some aspen and jack pine stands.

Stage 3. Understory reinitiation.
At this point, a stand undergoes demographic transition from one cohort to more than one cohort. There may be a wave of high mortality as many trees reach old age at the same time. The crowns of the trees are now large enough so that when one dies, the surrounding trees cannot fill the gap. As a result, a new cohort of trees has space to enter the canopy. The diameter distribution becomes a compound of the two cohorts - an old unimodal peak in larger size classes and a new peak in the small size classes.

If the stand was originally composed of a pioneer species (e.g., paper birch, aspen or yellow poplar), shade-tolerant trees such as sugar maple or beech may begin entering the canopy. If there are more gaps in the canopy and more light on the forest floor, some of the mid-tolerant trees, such as white ash, red maple, yellow birch and white pine, also may enter the canopy. Mortality undergoes a transition from mostly density-dependent self-thinning to mostly density independent mechanisms, such as senescence, windthrow (due to weakened wood caused by heartrot) or disease. The stand begins to take on "old growth" characteristics, with large rotten logs on the forest floor, many tree sizes and an uneven canopy surface.

Stage 4. Old, multi-aged community.
At this point, demographic transition is complete; the forest has many age classes and size classes of trees in the canopy. There may be few or no remnants left from the original cohort. Mortality is continuous at a relatively low level, with death occurring mainly in individuals or small groups of trees.

In the Eastern Deciduous forest region, this model of forest regeneration may also apply to communities of mid-tolerant species where seed sources of shade-tolerant species do not exist.

This situation is most common in the so-called Oak-hickory region where potential shade-tolerant, climax dominants such as sugar maple, American beech and perhaps basswood, are believed to have been eliminated from much of the landscape by wild- and human-caused fires in the pre-European settlement period. Under these conditions, mixed oakhickory forests developed because oaks and hickories have a strong ability to sprout when tops are killed by fire. Today, however, when wildfires no longer control understory competition, these forests have almost no oak regeneration. This is especially so on the most productive mesic sites. Instead, oak-dominated forests tend to be replaced by species of moderate shade tolerance, such as red maple, red elm, boxelder or shagbark hickory. In the absence of management, this type of forest tends to follow the cycle described above: stem exclusion, to understory reinitiation, to old, multiaged community.

Other factors influencing change in composition and structureOther factors influencing change in composition and structurev

Composition of original stand
Stand composition prior to disturbance or silvicultural manipulation has a strong influence on the direction and rate of change. Some ecologists refer to this effect as "biological legacy." The most obvious influences are factors such as the abundance and composition of advance reproduction, the ability of member tree species to sprout and seed availability, but other variables are also important. These include the condition of the forest floor itself (e.g., exposed mineral soil, type of humus, moss) and the presence of a vigorous herbaceous or shrub layer that may compete with tree seedlings. Also important may be population levels of various herbivores, including seed predators, as well as soil microorganisms, both pathogenic and beneficial.

Seed source availabilityAvailability of seed is an obvious requisite for any compositional change. In most forests, species composing the existing stand supply the vast majority of seed that reaches the forest floor. However, seed presence on the forest floor does not necessarily guarantee species' success in an germination and survival. Oak stands, especially on mesic sites, are good examples of such a condition. Often wind- and bird-disseminated seeds from sources outside of the stand comprise the majority of successful reproduction. Because availability of such external seed sources can be observed only on a case-by-case basis, no general predictions of successional change in any given forest cover type can be made. Thus, all generalized successional models are predicated on seed source availability.

Disturbance
Compositional and structural changes are also strongly affected by disturbance. Not only is the type of disturbance important (e.g., fire, logging, windthrow), but so is the intensity and timing of the disturbance. We often hear generalized statements, such as: "Disturbance, like windstorms, fire or clearcutting, causes pioneer species to take over the site." The three types of disturbance mentioned in this statement differ not only in the manner in which they affect the canopy, but also in how they affect many other factors, such as seedbed conditions, seed sources and response of competing vegetation. For example, low-severity disturbance (e.g., cutting during the winter with no disturbance of ground or advance regeneration) would favor shade-tolerant species. Medium severity disturbance (e.g., cutting in the summer with some soil scarification, or windthrow that creates tip up mounds) would likely lead to a mixture of shade-tolerant and mid-tolerant species. And finally, high-severity disturbance (e.g., hot fire that consumes organic matter on the forest floor as well as the canopy) would create an opportunity for invasion of intolerant pioneer species.

Any of the above factors may play a role in the dynamics of a given stand or vegetation community, making specific predictions of change difficult, if not impossible. However, forest managers must understand the effects of these factors as they apply in their region.