2.2. Stand Dynamics: Disturbances


Disturbances influence and alter many key ecosystem processes and structures that occur during stand dynamics. An understanding of disturbances informs our knowledge of how a forest will change over time. Most silvicultural treatments are in fact disturbances, so a working knowledge of disturbance ecology can be directly applied to understanding the effects of management actions.

Picket and White (1985) define a disturbance as any relatively discrete event in time that disrupts ecosystem, community, or population structure and changes resources, substrate availability, or the physical environment.

Discrete indicates that a disturbance can be bounded in both space and time. For example, wind may last a few minutes and fell a single tree as in a thunderstorm, or it may last days and damage timber over thousands of square miles, as in a major hurricane. Both events can be clearly described in terms of when and where they occurred.

All disturbances are stochastic. Stochastic is a term used to define an event that will occur in the future, but it is not known when, where, or how severe it will be. It is almost certain that a hurricane will strike North America in the next hurricane season, but it is currently not known when (June, August, October?), where (New England, Florida, Mexico?), or how severe it will be (Category 2, 3, 4?). While stochastic events cannot be described with certainty, they can be explained probabilistically. For instance, based on past observations, we know that hurricanes are most likely to occur in early September, and rarely occur between January and April (see http://www.nhc.noaa.gov/climo/#cp100). So while a disturbance may be stochastic, we are often still able to quantify our uncertainty and risk for a specific silvicultural activity in a known location to some extent. We are also able to use this information to predict how a specific forest is likely to change over time, at least in general terms.

Disturbances vary widely in their nature. Here is a list of but a few examples of disturbances that have affected forest ecosystems:

  • Mount Saint Helens erupting (1980)
  • Siberian meteor strike (1908)
  • Glaciers
  • Landslides or avalanches
  • Floods
  • Drought
  • Wildfires in Yellowstone (1988)
  • Fine-scale wind storms
  • Tornadoes
  • Ice storms
  • Gypsy moth
  • Southern pine beetle
  • Herbivory

Describing Disturbances

Disturbances may be quantified and described by a variety of commonly used metrics. Picket and White (1985) describe nine metrics that are most commonly used:

  • distribution,
  • frequency,
  • return interval,
  • rotation period,
  • predictability,
  • area or size,
  • intensity,
  • severity,
  • and synergism.

Distribution is the spatial pattern of a disturbance, and indicates whether it is patchy or clumped, and how it is arranged with respect to site resources (Picket and White 1985).

Figure 2.2.1. In 1908 a meteor struck near Tunguska, in Russian Siberia, decimating more than 800 square miles of spruce-fir forests. The distribution of this disturbance was relatively homogenous over this large area, as there was enough force from the blast to kill almost every tree in the area regardless of differences in soils, topography, or stand structure. Photo Credit: Public domain, taken during the 1927 Kulik expedition

Figure 2.2.2. A windblown spruce in Slovakia has created a small tip-up mound. Fine-scale windthrow is a very patchy, heterogeneously distributed disturbance that occurs more in shallow-rooted species. Tip-ups occur more frequently on low-strength or shallow soils, such as those found in seeps. Photo Credit: Milan Zubrik, Forest Research Institute - Slovakia, Bugwood.org

Frequency is the number of events per time that occur in a specific area (Picket and White 1985). When describing frequencies it is important to note the area, as frequencies differ geographically. For example, 8 major hurricanes struck the six-state New England area in the roughly 300 year period between 1635 and 1938 (Foster et al. 2004). Thus, the historical frequency of major hurricanes in New England is 8 / 303 = 0.026 per year. This can be more intuitively expressed as 8 / 3 = 2.6 hurricanes per century. This frequency is a statistical mean over that time period. While there were only 2 major hurricanes in the 1600's and 1700's combined, there were 5 in the 1800's. There will always be variability about the mean. Knowing this variability is important when making statistical predictions about the likelihood of future disturbances affecting an area.

Return interval is the inverse of frequency, or the mean time between events that occur in a specific area (Picket and White 1985). Using our example above, the return interval for hurricanes in New England would be 303 / 8 = 38 years. As with frequencies, this is only a mean. Due to the stochastic nature of hurricanes, it is highly unlikely that a major hurricane will strike exactly every 38 years. Generally speaking, the more intense the disturbance, the longer its return interval will be. This trend is evident in return intervals for hurricanes striking the eastern US Coast.

Figure 2.2.3. Estimated return interval in years for hurricanes within 50 nautical miles of the coast with wind speeds in excess of 64 knots. Note that these return intervals are shorter than for the more intense hurricanes shown on the next map. Map Credit: National Hurricane Center 2012

Figure 2.2.4. Estimated return interval in years for hurricanes within 50 nautical miles of the coast with wind speeds in excess of 96 knots. Note that these return intervals are longer than for the less intense hurricanes shown on the previous map. Map Credit: National Hurricane Center 2012

Rotation period is the time it takes one type of disturbance to impact an entire area (Picket and White 1985). For example, if 10% of a stand is felled by wind each decade as is common in many northern and central hardwood stands, then it will take 100 years for the entire stand to be disturbed by wind on average (Runkle 1982). One hundred years is the rotation period. Due to the stochastic variability of disturbances, it is highly unlikely that every single tree will actually be disturbed within a single rotation period. Rather species resistant to windthrow rooted on high-strength, deep soils may survive undisturbed for 100's of years, while more susceptible species rooted in low-strength, shallow soils may be felled by wind every 50 years. Rotation period is thus describing the mean behavior of a disturbance in a simplistic way, and does not adequately capture the variability involved. Variability is better addressed by predictability.

Predictability is a scaled inverse function of variance in the return interval. In plainer language, the more variable the return interval is, the less predictable the disturbance is. Looking at our previous examples, compare a meteor strike to fine-scale windthrow that tips up a single tree. For a given 100 acres of forest, we have a much better chance of accurately predicting the proportion of the stand that will be disturbed by fine-scale windthrow over the next decade than we do of determining the likelihood that a meteor will strike the stand. This is because there is less variability associated with windthrow compared to meteor strikes.

Area or size is some measure (acres, square miles, etc.) of the area disturbed for a specific event or time period.

Figure 2.2.5. This southern pine beetle spot covers an area of less than an acre. If promptly treated it is often possible to prevent the beetles from spreading to the surrounding stand. If left untreated, this spot might expand to thousands of acres in size under the right conditions. Photo Credit: Southern Archive, USDA Forest Service, Bugwood.org

Intensity is a measure of the force of a disturbance over a certain area and time. For example, the intensity of a fire can be described in BTU's / acre / hour, where BTU (British thermal unit) is a measure of heat. In this example, a fire with more BTU's per acre per hour would be more intense, and thus hotter. Intensity is a characteristic of the disturbance, and does not necessarily describe the impact that disturbance has on a particular ecosystem. Stand structure, topography, and other factors will also play a role in the effect of a disturbance. While not directly related to impact due to these confounding factors, intensity is nonetheless generally positively correlated to impact. More intense disturbances have the potential to do more damage than less intense disturbances for a given ecosystem.

Figure 2.2.6. A relatively intense (high BTU / acre / hour) lightning ignited wildfire in Oregon. Photo Credit: Dave Powell, USDA Forest Service, Bugwood.org

Severity is a measure of the impact a disturbance actually has on an ecosystem. Severity can be characterized in many ways, including percent mortality, basal area affected, tons of soil eroded, etc. Severity differs from intensity in that it describes the effect of the disturbance, not the force required to achieve that effect.

Figure 2.2.7. Aftermath of the severe 1988 wildfires in Yellowstone National Park, Wyoming. Quantified in terms of mortality, the effect of the fire was severe in this lodgepole pine stand, with nearly 100% mortality. Photo Credit: Billy Humphries, Forest Resource Consultants, Inc., Bugwood.org

A diagram depicting the inverse relationship between disturbance severity and frequency using examples from this section.

Figure 2.2.8. Disturbance frequency and severity tend to be inversely related. Less severe disturbances occur more frequently. This is fortunate for foresters and the general populace, as it is the reason that more severe hurricanes, tornadoes, droughts, and other disturbances occur less frequently than less severe and damaging events.

Synergism is the effect one disturbance may have on the probability of another disturbance occurring (Picket and White 1985). For example, beetle-killed spots within a loblolly pine stand may be relatively small in proportion to the stand. However, they are areas of high fuel loading, increasing the risk that a lightning-ignited wildfire will start in these areas and spread to the surrounding stand. This effect is exacerbated further during an extreme drought, such as the drought in Texas from 2010-2011 that contributed to the worst wildfire season (4+ million acres burned) in state history.

Disturbances and Stand Dynamics

Top-Down versus Bottom-Up

A disturbance can be classified by the effects it has on stand structure as much as by the characteristics of the disturbance itself. Some disturbances, such as a surface fire, act from the bottom of a stand up. They influence herbaceous and shrub strata most, sometimes without having a substantial effect on the overstory. Other disturbances, such as an ice storm, act from the top of a stand down. These disturbances break large branches and tops out of trees. They may or may not release the midstory or understory strata, depending on the percentage of the canopy disturbed and the ability of the overstory trees to expand their crowns and fill the newly formed gaps. While these general descriptions describe how a disturbance acts on a stand, they do not tell the whole story. Fires vary in severity, from low severity surface fires that have little impact on the overstory, to high severity crown fires that result in mortality of most of the overstory. The utility of this simplistic classification system is in understanding what structures within a stand are most likely to be affected by a particular disturbance, and thus what effect that disturbance may have on stand dynamics.

Figure 2.2.9. A prescribed fire in a 12-year-old slash pine stand in south Florida. This surface fire is a classic bottom-up disturbance. It is being used as a silvicultural tool to control the dense understory dominated by saw palmetto and gallberry, while having little if any negative impact on the overstory. If the fire results in high mortality in the understory, the slash pines will have greater access to nutrients and water following the fire, and should grow more rapidly. Photo Credit: Dale Wade, Rx Fire Doctor, Bugwood.org

Figure 2.2.10. Snow and ice disturb an alder stand in the Czech Republic. Ice damage usually breaks large limbs and causes crown damage, but does not itself cause significant mortality. This top down disturbance alters resource availability, allowing more light to the understory and releasing nutrients from felled branches as they decompose. This may result in partial release of a new cohort or in understory reinitiation. Photo Credit: Petr Kapitola, State Phytosanitary Administration, Bugwood.org

Resistance versus Resilience

Similar disturbances may have dissimilar impacts on the structure of two different stands. The structure of a stand, silvics of the dominant species, soils, and topography all dictate the impacts of a disturbance on a stand. For example, if two identical wind events occur in two stands with different species, one stand may be severely impacted, while another is not disturbed at all. The depth of rooting and soil strength will be important variables that determine what percentage of trees tip up. The wood properties including density and modulus of elasticity, stem taper, and crown size will play a role in whether trees snap, and if so where they snap. These properties vary not only by species, but with the age of a stand; a young stand may not be affected in the same way as an older stand by identical disturbances.

Depending on stand structure, the silvics of the species, and the nature of a disturbance, forested ecosystems may be either resistant or resilient.

Resistant ecosystems show little impact due to repeated disturbances over time. However, if disturbances become too intense, structure and function may be severely impacted. At this point a resistant ecosystem does not easily recover.

By contrast, a resilient ecosystem is often immediately impacted by even low intensity disturbances. However, it has the capacity to quickly recover structurally and functionally to levels approaching the pre-disturbance condition.

These patterns have been observed in tree species in response to hurricanes (Boucher et al. 1994). Some species are resilient, experiencing nearly 100% mortality, yet quickly recovering through the establishment of a dense new cohort. Others experience almost no hurricane-related mortality, but may be impacted by other stressors, such as competition, and may not successfully regenerate over time. The silvics of the species in question are as important to determining the stand dynamics as are the characteristics of the disturbance.

A diagram depicting the difference between resistant and resilient ecosystems.

Figure 2.2.11. A simplistic representation of how ecosystem function responds to disturbances in resistant versus resilient ecosystems.

Another example is the effect of fire on trembling aspen contrasted with coastal redwood. Trembling aspen is a resilient forest cover type. Fire may cause 100% mortality of the overstory, but aspen quickly sprouts from the roots, producing a new, rapidly-growing stand that can outcompete other species. Coastal redwood represents a resistant forest cover type. The bark of redwoods is extremely thick, and they are able to withstand relatively intense fires without experiencing significant mortality. In part because they are adapted to resist disturbances, redwoods can live for 1000's of years. Individual redwood trees may live for centuries longer than individual trembling aspen stems, but both are forest cover types that can remain on the landscape for millennia in the face of repeated disturbance by fire.

Beyond the silvics of the dominant tree species in a stand, availability of nutrients also can dictate whether an ecosystem is resistant or resilient (Herbert et al. 1999). Stands treated with fertilizer tend to have larger trees that are more subject to wind damage during hurricanes. However, while damage may be greater, increased nutrient availability leads to more rapid recovery from disturbance as trees can add new leaf area more quickly. Thus, the silvicultural addition of nutrients can shift an ecosystem from resistance toward resilience (Herbert et al. 1999).

The nature of the disturbance itself also dictates the extent to which forests are either resilient or resistant. More intense disturbances generally cause more damage. While a stand may be resistant to a low intensity disturbance, a high intensity disturbance may cause significant damage. In some cases, this allows for the establishment of new species better suited to the post-disturbance environment (Halpern 1988). As long as the new species are not undesirable invasive species, the shift in composition brought on by disturbance is a natural pathway of forest succession. While the structure of the stand may change through succession, if it recovers similar ecosystem functions post-disturbance, then the forest community remains resilient in the face of disturbance.

Spatial Variability and Disturbance

The interactions of disturbances with stand structure create complex responses that are dependent on many variables.

  • Effects of disturbances on stand structure and function depend on:
    • age and age class structure,
    • abundance of competing vegetation,
    • density,
    • canopy height and crown morphology,
    • stand history,
    • tree vigor and health,
    • soils and topography, and
    • tolerance of species to disturbance (windfirmness, drought tolerance, etc.).
  • Effects on regeneration following disturbances depends on:
    • mode of reproduction (sexual, vegetative),
    • propagule dispersal distance,
    • seed bank contents, and
    • distance from adjacent stands.

These interactions become increasingly complex when their spatial distribution is also considered. Most disturbances will create spatial variability in a stand as a product of both the patchy nature of the disturbance itself, and the structural variability inherent to forest ecosystems. For example, fire often varies in its intensity due to differences in fuel loading, topography, and weather, and thus tends to alter the abundance of favorable microsites (small areas) where regeneration can succeed (Oswald and Neuenschwander 1993).

Gaps formed by fine-scale wind disturbances vary in their size, and may increase in size as a result of subsequent wind events (Curzon and Keeton 2010). While smaller gaps favor regeneration of shade tolerant overstory species, larger gaps may also favor the regeneration of shade intolerant pioneer species (Runkle 1981, 1982).

A diagram depicting spatial variability in light and root competition in a gap.

Figure 2.2.12. The spatial pattern of habitats within gaps is complicated by interactions between light availability and belowground competition. More radiant energy from light reaches the north side of gaps in the northern hemisphere, since the sun is in the southern portion of the sky. The southern edge of gaps is more shaded, creating different microsites more or less suitable to regeneration of different species, depending on their shade tolerance. Shade tolerant species are better suited to the south edge of gaps, while shade intolerant species are more adapted to the center or the north edge of gaps. Couple this with more intense belowground competition for water and nutrients from the adjacent stand near the edge of gaps, and a spatially complex resource environment emerges in each gap following disturbance by wind.

The balance between belowground competition for nutrients and water, and aboveground competition for light varies not only within a gap, but also across different sites. In gaps on wetter sites with greater nutrient availability, regeneration tends to be almost exclusively limited by light availability. This trend is further compounded by the fact that leaf area, and thus shade, tends to be highest on sites with high nutrient and water availability. By contrast, gaps on more xeric sites with lower nutrient availability tend to be limited by belowground competition and not light (Coomes and Grubb 2000). Leaf area and the shade it casts tends to be less on these sites due to nutrient limitations.


Figure 2.2.13. Mass wasting on Macomb Mountain in the Adirondacks of upstate New York. This landslide removed not only all the vegetation, but also severely disturbed the already thin topsoil. Photo Credit: Jeremy Stovall

Figure 2.2.14. A large gap is blown down in an Engelmann spruce stand in Colorado. Photo Credit: William Ciesla, Forest Health Management International, Bugwood.org

Figure 2.2.15. The Mississippi River is flooded out of its banks. Flooding prevents the roots from acquiring the oxygen that they need to respire and survive. Species vary in their tolerance to flooding in the root zone, which is why many species are never found on soils that flood seasonally. Photo Credit: Brian Lockhart, USDA Forest Service, Bugwood.org

Figure 2.2.16. On May 18, 1980, Mount Saint Helens in Washington state erupted with such force that it destroyed more than 125,000 acres of forest land. Despite the destruction, the forests are regenerating, although changes in composition have occurred. Photo Credit: Joseph O'Brien, USDA Forest Service, Bugwood.org

Figure 2.2.17. On May 18, 1980, Mount Saint Helens in Washington state erupted with such force that it destroyed more than 125,000 acres of forest land. Despite the destruction, the forests are regenerating, although changes in composition have occurred. Photo Credit: Joseph O'Brien, USDA Forest Service, Bugwood.org

Figure 2.2.18. Larvae of the gypsy moth have defoliated many of the hardwood species on a ridge in Virginia as seen in summer. A single defoliation in the absence of other severe disturbances typically does not result in substantial mortality. However, if gypsy moth outbreaks occur two years in a row, many trees are unable to recover and die. Outbreaks are inconsistent from year to year and are difficult to predict. Photo Credit: Tim Tigner, Virginia Department of Forestry, Bugwood.org


Boucher, D. H., J. H. Vandermeer, M. A. Mallona, N. Zamora, and I. Perfecto. 1994. Resistance and resilience in a directly regenerating rainforest: Nicaraguan trees of the Vochysiaceae after Hurricane Joan. Forest Ecology and Management 68:127-136. http://dx.doi.org/10.1016/0378-1127(94)90040-X

Coomes, D. A. and P. J. Grubb. 2000. Impacts of root competition in forests and woodlands: A theoretical framework and review of experiments. Ecological Monographs 70:171-207. http://dx.doi.org/10.1890/0012-9615(2000)070%5B0171:IORCIF%5D2.0.CO;2

Curzon, M. T. and W. S. Keeton. 2010. Spatial characteristics of canopy disturbances in riparian old-growth hemlock – northern hardwood forests, Adirondack Mountains, New York, USA. Canadian Journal of Forest Research 40:13-25. http://dx.doi.org/10.1139/X09-157

Foster, D. R., G. Motzkin, J. O'Keefe, E. Boose, D. Orwig, J. Fuller, and B. Hall. 2004. The environmental and human history of New England. Pages 43-100 in D. R. Foster and J. D. Aber, editors. Forests in time: The environmental consequences of 1,000 years of change in New England. Yale University Press, New Haven, CT. ISBN: 0300092350

Halpern, C. B. 1988. Early successional pathways and the resistance and resilience of forest communities. Ecology 69:1703-1715. http://dx.doi.org/10.2307/1941148

Herbert, D. A., J. H. Fownes, and P. M. Vitousek. 1999. Hurricane damage to a Hawaiian forest: Nutrient supply rate affects resistance and resilience. Ecology 80:908-920. http://dx.doi.org/10.1890/0012-9658(1999)080%5B0908:HDTAHF%5D2.0.CO;2

National Hurricane Center. 2012. Tropical cyclone climatology: Hurricane return periods. National Oceanic and Atmospheric Administration, National Weather Service, National Hurricane Center, Miami, FL. Last accessed 24 July 2012. http://www.nhc.noaa.gov/climo/

Oswald, B. P. and L. F. Neuenschwander. 1993. Microsite variability and safe site description for western larch germination and establishment. Bulletin of the Torrey Botanical Club 120:148-156. http://dx.doi.org/10.2307%2F2996944

Pickett, S. T. A. and P. S. White, editors. 1985. The Ecology of Natural Disturbance and Patch Dynamics. Academic Press, Inc., Orlando, FL. ISBN: 0125545215

Runkle, J. R. 1981. Gap regeneration in some old-growth forests of the Eastern United States. Ecology 62:1041-1051. http://dx.doi.org/10.2307%2F1937003

Runkle, J. R. 1982. Patterns of disturbance in some old-growth mesic forests of Eastern North America. Ecology 63:1533-1546. http://dx.doi.org/10.2307/1938878