2.3. Stand Dynamics: Stand Structure


An understanding of various stand structures and the processes that create them are critical to your understanding of stand dynamics. Stand structure simply refers to any physical aspect of the forest, including live trees, standing dead trees, fallen dead trees, other associated non-arboreal vegetation, roots, the litter layer, and the mineral soil. We use a number of different attributes to describe and characterize each of these structures. For example, live trees may be characterized based on size (quadratic mean diameter, height, volume), density (trees per acre, basal area, stand density index), stocking (relative density), composition (species), or crown morphology (live crown ratio, crown width). Dead trees may be quantified in similar terms, and may also be characterized by their state of decay and mode of death (windthrow, fire, insects or disease, drought, flood, herbivory, etc.). When walking through the woods, be observant of the various stand structures around you, and think about the processes that may have caused them. Then, whenever you read about or try to describe stand structures, you can remember stands you have seen with similar characteristics.

To simplify our understanding of stand dynamics, we will use a four-stage model developed by Oliver and Larson (1996). Remember that a model is merely a simplified version of a system. This particular model is useful because it is simple and easy to remember, yet is widely applicable to the majority of forest types in the world. The OL model is characterized by four stages. The first three are based on individual key processes that separate them from one another. The fourth stage, old growth, includes a wide variety of different processes that can occur for centuries, and describes a structural condition more so than one key process. Each stage is described below, along with key structures, processes, and examples illustrated with photos.

Oliver and Larson Model

  • Stand initiation
  • Stem exclusion
  • Understory reinitiation
  • Old growth

Stand Initiation


New individuals and species continue to appear for several years following a disturbance. Stands are young and can be quite dense. A stand initiating disturbance marks the beginning of this stage, while canopy closure marks the end (Oliver and Larson 1996).

Key Structures

Adapted from Franklin et al. (2002)
  • Live trees
  • 100% live crown ratios
  • Debris or slash
  • Legacy trees (live or dead)

Key Processes

Adapted from Franklin et al. (2002)
  • Stand initiating disturbance(s)
  • Establishment of new cohort
    • Colonization by new seed
    • Germination from seed bank
  • Minimal or no nutrient limitations
  • Rapid growth


Figure 2.3.1. Silvicultural activities may mimic (albeit never exactly) some natural disturbances, and may actually serve as the stand initiating disturbance. Photo Credit: Jeremy Stovall, clearcut in a Virginia mixed-pine hardwood stand on a seasonally flooded site.

Figure 2.3.2. Legacy trees are those left from a previous rotation. They will be very apparent when the new stand is relatively young and short-statured by comparison. While this photo shows a live tree, legacy structures can also include standing dead trees called 'snags'. Photo Credit: Jeremy Stovall, legacy eastern hemlock in a hemlock-northern hardwoods stand in the Adirondack Mountains of New York.

Figure 2.3.3. In an older stand legacy trees may be identified based on the large, poorly pruned lower limbs and the fact that they are generally larger in diameter than the other trees, like the tree in the center of this photo. This is an indication that the legacy tree developed when the site was a pasture, as it would have had more available light and would not have self-pruned well. These trees are commonly referred to as ‘wolf trees’ since they use a higher proportion of the site’s resources. Photo Credit: Jeremy Stovall, a wolf-tree in a 65-year-old naturally seeded loblolly pine stand in East Texas.

Figure 2.3.4. During stand initiation trees tend to be small and thus have ample light, even close to the ground. Live crown ratios, or the percentage of the tree's height covered with live foliage, are thus usually 100%, and lower branches have not yet started to self prune. Photo Credit: Jeremy Stovall, a 3-year-old varietal loblolly pine stand planted near Summerville, South Carolina.

Figure 2.3.5. A large amount of slash, or woody debris, on this recently clearcut site is a typical feature of the stand initiation stage. Slash contains nutrients, such as nitrogen and phosphorus, that are released as it decomposes. Because there is usually ample slash, and trees remain small, significant nutrient deficiencies do not typically develop until after stand initiation. Photo Credit: Jeremy Stovall, a recently clearcut hardwood stand being bedded for conversion to a pine plantation in East Texas.

Stem Exclusion


After several years (3 to 50, depending on location and species), new individuals stop appearing and already established individuals begin to die. Those that survive grow and begin to manifest differences in height and diameter. Species may dominate the site for a time before another species takes over. Stands are very dense, and still relatively young (Oliver and Larson 1996). Canopy closure marks the onset of this stage, while density independent mortality, gap formation, and significant development of new vegetation in the understory signals the end. Both stand initiation and stem exclusion phases may be described as ‘early successional’ since they occur early in succession.

Key Structures

Adapted from Franklin et al. (2002)
  • Less than 100% live crown ratios
  • Vertically differentiated canopy
  • Heavily shaded understory

Key Processes

Adapted from Franklin et al. (2002)
  • Canopy closure
  • Density DEPENDENT mortality
  • Competitive exclusion of understory
  • Crown differentiation
  • Lower canopy tree loss
  • Self pruning
  • Nutrient limitations develop


Figure 2.3.6. Stands in the stem exclusion stage are typically very dense, with heavily shaded understories that usually lack significant shrub or herbaceous species. Because of the high density, many of the smaller trees are being shaded out by the larger ones, eventually leading to their death. Because the high density is causing the death of the smaller trees, this is referred to as density dependent mortality. Photo Credit: David Moorhead, UGA, Bugwood.org, precommercial thinning is needed on this overstocked loblolly pine stand in Arkansas.

Figure 2.3.7. The nearly completely shaded understory in this hardwood stand is the result of canopy closure, or the point at which trees grow large enough that their crowns touch. Once this occurs self pruning also begins. As lower branches become more shaded, they no longer produce enough sugar via photosynthesis to support themselves. The tree's response is to shed, or prune, the branch, resulting in a clear main stem or bole that is of higher value as sawtimber or veneer since it has fewer knots. Photo Credit: Jeremy Stovall, a 15-year-old mixed species bottomland hardwood stand in East Texas.

Figure 2.3.8. During stem exclusion weaker individuals are gradually shaded out while stronger individuals overtop them. This process is crown differentiation. The process of crown differentiation is most simply explained in a stand comprised of a single species, since crown morphology and growth rates are very similar among all trees. Trees that grow slightly faster due to either better genetics, a more favorable microsite, or chance events become either dominant (dom) or codominant (co). Slower growing trees end up either being completely shaded from the sides like intermediates (int), or are completely shaded from the sides and from above like suppressed or overtopped trees (ovt). Eventually these weaker individuals will succumb to density dependent mortality in most species (like the dead tree top-right). Photo Credit: Michael Fountain, a slash pine plantation.

Understory Reinitiation


Later in stand development forest floor herbaceous species, shrubs, and advanced regeneration all again appear and survive in the understory, although growth may be slow (Oliver and Larson 1996). They are able to do so because large dominant trees are dying due to insects, disease, lightning, windthrow, or other causes, forming canopy gaps and freeing up site resources. Density independent mortality, gap formation, and significant development of new vegetation in the understory indicates the beginning of this stage, while pioneer cohort loss, substantial horizontal and vertical variability, and a high density of large diameter, often decadent trees roughly correspond to the end.

Key Structures

Adapted from Franklin et al. (2002)
  • Understory herbaceous layer
  • Shade tolerant cohort
  • Few smaller canopy gaps
  • Standing dead trees
  • Some large woody debris
  • Some uprooted or snapped trees

Key Processes

Adapted from Franklin et al. (2002)
  • Density INDEPENDENT mortality
  • Canopy gap initiation
  • Understory redevelopment
  • Establishment of shade tolerant species
  • Maturation of pioneer cohort
  • Canopy elaboration
  • Nutrient limitations persist but lessen


Figure 2.3.9. One of the key processes of understory reinitiation is density independent mortality, or the death of trees due to factors unrelated to competition with other trees. Such processes might include attack by insects or disease, windthrow, or lightning strikes such as the tree seen in this photo. Density independent mortality is important because it initiates the formation of canopy gaps, which free up light, water, and nutrients that allow a new cohort of shade tolerant species to establish. These newly available resources also allow for the establishment and growth of a vigorous herbaceous strata. Photo Credit: Jeremy Stovall, a burned snag in a 70-year-old longleaf pine stand in East Texas.

Figure 2.3.10. Large trees snapped or tipped up by windthrow are another mechanism of canopy gap formation. When the wood from this snapped tree eventually falls to the ground, it will decompose, releasing nutrients that will be available to other vegetation. As the stand becomes more spatially patchy and organic matter accumulates both in the leafy litter layer and due to coarse woody debris, nutrient deficiencies lessen to some extent. Photo Credit: Jeremy Stovall, snapped yellow birch in the Adirondack Mountains of New York.

Figure 2.3.11. Dense herbaceous and shrub layers are formed during understory reinitiation in response to increasing resource availability. Photo Credit: Jeremy Stovall, a 70-year-old longleaf pine stand in East Texas.

Figure 2.3.12. Shade tolerant hardwoods are able to establish in previously pure pine-dominated stands in the understory reinitiation stage across much of the southern US. Because pines are shade intolerant they are not always able to regenerate beneath themselves. Thus pines will eventually be replaced by mixed hardwood species such as red maple, white oak, and blackgum on this more mesic site. Photo Credit: Jeremy Stovall, a 65-year-old lobolly pine - mixed hardwood stand in East Texas.

Old Growth


Much later in the life of a stand, overstory trees die in a spatially patchy pattern while some of the younger trees in the understory begin to grow into the overstory (Oliver and Larson 1996). In most forest types old growth may be roughly distinguished from understory reinitiation by the loss of the pioneer cohort of trees. However, the boundary between understory reinitiation and old growth is the most gradual and difficult to discern of any in the OL model. The presence of a preponderance of the structures and processes listed below should be used as the best indicator of an old growth forest. Old growth is merely a descriptor of stand structural condition. It has no bearing on what human-caused activities or disturbances may have occurred in the past. It is thus a more precise term than 'virgin' forest, which indicates a lack of human intervention, but may not necessarily correlate to old growth stand structures due to the influence of natural disturbances. For instance, a virgin forest leveled in a hurricane might only be 30 years old, while an old growth forest could conceivably (although uncommonly) exist on a site that was previously used for agriculture, but abandoned in the 1600's. Both old growth and understory reinitiation phases may be described as ‘late successional’ since they occur later in succession.

Key Structures

Adapted from Franklin et al. (2002)
  • Large diameter live trees
  • Large branches
  • Rich epiphyte community
  • Continuous vertical foliar profile
  • More standing dead trees
  • More large woody debris
  • More uprooted or snapped trees
  • Horizontally patchy forest
  • Large gaps
  • Densely regenerating old gaps

Key Processes

Adapted from Franklin et al. (2002)
  • Canopy gap expansion
  • Uprooting and snapping of large trees
  • Live tree decadence (poor form and disease)
  • Development of large branches
  • Pioneer cohort loss
  • Nutrient limitations decline as organic matter accumulates


Figure 2.3.13. On shallow or wetter soils trees can tip up large volumes of soil when they are felled by wind disturbances. These tip up mounds can be quite large when multiple trees are blown down at once. The soil mound in the middle of a newly formed gap forms a favorable microsite for the establishment of new tree species. Over time as the soil continues to erode these trees may find themselves with little to support their roots, and thus they may not survive. The excavated pit also forms a favorable, wetter microsite for various wildlife species like amphibians. In parts of the country, it is easy to identify land that was never plowed for agriculture based on the undulating microtopography caused by tip up mounds occurring periodically over centuries. Photo Credit: Jeremy Stovall, a large tip up mound caused by the blow down of several large diameter eastern hemlocks in the Adirondack Mountains of New York.

Figure 2.3.14. Dead wood is a very common feature of old growth forests throughout the world. Dead wood may be standing (snags), or fallen (coarse woody debris, CWD). Dead wood forms important habitat structures for many species of wildlife and stores nutrients that are gradually released to surrounding vegetation. Photo Credit: Jeremy Stovall, a tip up mound caused by the blow down of an eastern hemlock in the Adirondack Mountains of New York.

Figure 2.3.15. Canopy gaps vary in size depending on how they are formed. This is a hemispherical (i.e. fisheye) canopy photo of a relatively large gap formed by the blowdown of several large dominant trees near a stream. The lighted point indicates north. Gaps over small headwater streams can create well lighted patches more suitable to the growth of algae that serves as a food source to fish and aquatic invertebrates (Stovall et al. 2009). Photo Credit: Jeremy Stovall, a large canopy gap caused by windthrow in the Adirondack Mountains of New York.

Figure 2.3.16. Gap expansion is a process whereby an existing small gap grows larger through subsequent blow down events. When one tree is felled, it forms an edge in the canopy that makes it more likely that wind will be able to fell other trees. Many gaps in old growth forests are extremely patchy, often with some surviving large trees remaining intact in the gap, while many others are felled. Photo Credit: Jeremy Stovall, a large 2 acre canopy gap caused by a microburst (i.e. intense thunderstorm) in the Adirondack Mountains of New York.

Figure 2.3.17. In old growth riparian (stream-side) forests, large woody debris plays an important role in streams (Keeton et al. 2007). It forms habitat structures for fish and invertebrates, and has also been shown to alter the phosphorous cycle in small headwater streams (Valett et al. 2002). Photo Credit: Jeremy Stovall, large woody debris in a small headwater stream in the Adirondack Mountains of New York.

Figure 2.3.18. Because old growth forests have multiple cohorts, there are a range of tree sizes. Smaller trees occupy the understory, and there is typically a well developed midstory. While the overstory is patchy due to gaps, there remain many large trees. As a result, old growth forests tend have a vertically continuous foliar profile. In other words, there are leaves from ground level all the way up to the top of the tallest tree. This is the result of many different strata, or layers, of trees, not of any one dominant canopy tree having leaves all the way from top to bottom. Photo Credit: Jeremy Stovall, a canopy profile of an old growth hemlock - northern hardwoods stand in the Adirondack Mountains of New York.

Figure 2.3.19. Most people unfamiliar with the various stand structures of old growth forests associate the term with large trees. A high density of large diameter trees is indeed one of the best ways to identify an old growth stand. What defines a large tree will vary in different forest types. What may be quite large for a water oak or sugar maple may still represent a relatively small size for a western hemlock. Photo Credit: Jeremy Stovall, a large water oak (42 inches in diameter at breast height) in an old growth beech - magnolia forest in East Texas.

Figure 2.3.20. As trees become older and larger they often times become more susceptible to a range of minor disturbances. Non-lethal attacks by insects, heart-rot or other diseases, and breaking of large branches by wind or ice are all common occurrences that create large but poorly formed trees in old growth stands. All these factors combine to create a state called 'decadence'. Photo Credit: Jeremy Stovall, a large southern magnolia with characteristic decadent form in an old growth beech - magnolia forest in East Texas.


Franklin, J. F., T. A. Spies, R. V. Pelt, A. B. Carey, D. A. Thornburgh, D. R. Berg, D. B. Lindenmayer, M. E. Harmon, W. S. Keeton, D. C. Shaw, K. Bible, and J. Chen. 2002. Disturbances and structural development of natural forest ecosystems with silvicultural implications, using Douglas-fir forests as an example. Forest Ecology and Management 155:399-423. http://dx.doi.org/10.1016/S0378-1127(01)00575-8

Keeton, W. S., C. E. Kraft, and D. R. Warren. 2007. Mature and old-growth riparian forests: structure, dynamics, and effects on Adirondack stream habitats. Ecological Applications 17:852-868. http://dx.doi.org/10.1890/06-1172

Oliver, C. D. and B. C. Larson. 1996. Forest Stand Dynamics. update edition. John Wiley and Sons Inc., New York, NY. ISSN: 0471138339 https://elischolar.library.yale.edu/fes_pubs/1/

Stovall, J. P., W. S. Keeton, and C. E. Kraft. 2009. Late-successional riparian forest structure results in heterogeneous periphyton distributions in low-order streams. Canadian Journal of Forest Research 39:2343-2354.http://dx.doi.org/10.1139/X09-137

Valett, H., C. Crenshaw, and P. Wagner. 2002. Stream nutrient uptake, forest succession, and biogeochemical theory. Ecology 83:2888-2901. http://dx.doi.org/10.1890/0012-9658(2002)083%5B2888:SNUFSA%5D2.0.CO;2