3.4. Regeneration Methods: Shelterwood

Category: Silviculture Textbook Created: Monday, 20 August 2012 Last Updated: Tuesday, 18 September 2018

3.4. Regeneration Methods: Shelterwood

This chapter was authored by Lana Welford and Allie Williams, undergraduate students enrolled in FOR 347 during the Fall 2016 semester, as a bonus opportunity.  It has subsequently been edited by Jeremy Stovall.  The final version has been approved for posting by Lana and Allie.


The shelterwood method is a method of regenerating an even-aged stand in which a new age class develops beneath the moderated microenvironment provided by the residual trees. The sequence of treatments can include three distinct types of cuttings: (1) an optional preparatory cut to enhance conditions for seed production; (2) an establishment cut to prepare the seed bed and to create a new age class; and (3) a removal cut to release established regeneration from competition with the overwood. The shelterwood method is very similar to seed-tree regeneration but leaves a fuller overstory following the establishment cut (Adams et al. 1994).

  • shelterwood1_lockhart

Figure 3.4.1. A shelterwood three years after the establishment cut in a Nuttall and willow oak stand in the US South. Photo Credit: Brian Lockhart, USDA Forest Service, Bugwood.org

  • shelterwood2_katovich

Figure 3.4.2. A shelterwood following the establishment cut in an eastern white pine stand on the Menominee Indian Reservation in Wisconsin. Photo Credit: Steve Katovich, USDA Forest Service, Bugwood.org


While not as common as clearcutting in the US South, the shelterwood method is very common in some regions of the US and other parts of the world, and is perhaps the most adaptable and flexible of all even-aged regeneration methods. The shelterwood method is desirable to naturally regenerate a stand of intermediate to shade tolerant species, and can sometimes even be adjusted to accommodate some shade intolerant species. It is also ideal for heavy-seeded species where limited seed dispersal distances may prove challenging for the seed tree regeneration method.

Cuts Involved

There are primarily two types of cuts used in shelterwood systems, although a third is required occasionally:

Preparatory cuts are sometimes used to increase vitality and health of the intended residual trees in the stand. More often than not preparatory cuts are unnecessary due to prior thinning on a site. These cuts can be used to hasten the decomposition of the litter layer and to thin or remove less desirable species to prevent them from increasing in abundance once the site is open and competing with the crop species (Smith et al. 1997). Preparatory cuts may also be used to remove hazard trees if desired.

The establishment cut is used to open growing space in the canopy for establishment of regeneration in one operation to ensure uniformity. Ideally the site should be opened just enough to encourage establishment of the desirable species only (Smith et al. 1997). The best time for an establishment cut is during a good seed year after the seed has matured but before the seed germinates. The trees removed during the cut are generally made up of the least desirable trees and are in various crown classes including overtopped, intermediate, and occasionally some co-dominant. Trees with bad genes or poor quality should also be removed from the site (Hawley and Smith 1954). In some covertypes, abundant advanced regeneration is required prior to the establishment cut to achieve successful regeneration of the stand.

A removal cut is used to release the established seedlings. The object of the removal cut is to gradually or rapidly remove the old cohort as the newer cohort needs more growing space. Due to the nature of harvest operations, many individuals in the new cohort will be damaged or killed during this operation. It is thus essential to secure a high density of regeneration distributed throughout the stand prior to the removal cut to ensure enough members of the new cohort remain to adequately regenerate the stand. As with the seed-tree method, for a brief time between cuts a stand may be two-aged prior to harvest of the older cohort in the removal cut. However, this is a temporary condition during the regeneration process that persists for only a small proportion of the rotation, hence shelterwood being considered an even-aged regeneration method.


A diagram depicting the two cuts (establishment and removal) in the most common application of the shelterwood regeneration method .

Figure 3.4.3. The shelterwood regen method results in an even aged, single cohort stand upon completion of the removal cut. For a brief time between the establishment and removal cuts, the stand is two aged, with both the overwood cohort and the newly established cohort coexisting. However, because this is only a temporary condition, and the intent is the regeneration of an even aged stand, the shelterwood system is classified as even aged. This is a depiction of the most common application of the shelterwood regeneration method involving only two cuts (establishment and removal), with no preparatory cut.


A diagram depicting a three cut shelterwood, including the preparatory cut, the establishment cut, and the removal cut.

Figure 3.4.4. This is a depiction of a three cut shelterwood, including the preparatory cut, the establishment cut, and the removal cut.

Common Variations

There are many variations of the shelterwood regeneration method, allowing it to be modified to grow any species of trees except species that are extremely shade intolerant and those which are poorly suited to competing for belowground resources (Hawley and Smith 1954). The shelterwood is one of the most flexible techniques. While this is helpful for silviculturists in adapting to unique conditions found in each stand, it can lead to confusion in describing the precise nature of prescribed silvicultural treatments.

Uniform Shelterwood

In the uniform shelterwood method harvests are applied uniformly throughout the stand. There are three types of uniform shelterwoods.

  • The three-cut method involves a preparatory cut, an establishment cut, and a removal cut (Figure 3.4.4.).
  • The two-cut method involves an establishment cut and a removal cut only (Figure 3.4.3.). No preparatory cut is required, as the objectives described above for a prep cut are not needed on that stand. This is by far the most common application of the shelterwood regeneration method.
  • The one-cut method involves a removal cut only. This is done when a natural disturbance, such as blowdown, produces a regenerating cohort, thus making an establishment cut unnecessary.

Strip Shelterwood

In the strip shelterwood method cuts are applied in strips, rather than treating the stand uniformly.

A site is divided up into strips from one side of the site to the other and gradually the cuts are shifted across the site. In one year, an establishment cut will be made on the first strip. A few years later, a removal cut will be made in the first strip and an establishment cut will be made in the second strip. A few years after that, a removal cut will be made to the second strip, an establishment cut will be made to the third strip, and so on. While this is happening, the rest of the strips in the stand remain unharvested, unless intermediate treatments are required (Hawley and Smith 1954) (Figure 3.4.5).

There are many variations of the strip method. Within the strips themselves there can be narrower strips of preparatory, establishment, removal and final cuts done. Clearcuts have also been performed in alternating strips, while the strips in between the clearcuts provide protection and a seed source for natural regeneration. The strip shelterwood method can be varied, such as making strips of differing widths or orienting strips along the contour or other gradients rather than in a straight line (Hawley and Smith 1954).

Strip shelterwoods have advantages over the uniform method such as lower risk of windfall, more shade (depending on the directions of the strips compared to the sun), and less chance of damaging regenerated seedlings when removing mature overwood (Hawley and Smith 1954).

A diagram depicting a strip shelterwood.

Figure 3.4.5. The strip shelterwood is a modification of the uniform shelterwood allowing for repeated harvesting on a single stand along a series of long, linear strips.

Group Shelterwood

The group shelterwood method typically takes advantage of patches of existing regeneration, and works outward from those starting points to regenerate the stand. This is similar to strip shelterwood except it is used in the case of patches of regeneration in a stand caused by thinnings or disturbances such as insects, disease, or wind. The group shelterwood benefits more shade tolerant species while it is inappropriate for intolerant species or species sensitive to frost damage (Hawley and Smith 1954) (Figure 3.4.6.).

 A diagram depicting a group shelterwood.

Figure 3.4.6.  A group shelterwood.  Areas labeled 3 have either just had a removal cut performed, or had sufficient natural regeneration for other reasons (i.e. blowdown). An establishment cut has been completed in areas labeled 2, while areas labeled 1 are unharvested but mature. Areas labeled 4 are mid-rotation following establishment via the group shelterwood method, and will require appropriate intermediate treatments.
  • shelterwood_group_stovall1
Figure 3.4.7. An area with a high-density younger cohort in a group shelterwood, similar to areas 3 or 4 in Figure 3.4.6.  This bottomland hardwood stand on the Sherburne Wildlife Management Area in Louisiana was being managed for a generalized (i.e. diversity of taxa) wildlife objective. Photo Credit: Jeremy Stovall
  • shelterwood_group_stovall2
Figure 3.4.8. An area with a lower-density older cohort in a group shelterwood, similar to areas 1 or 2 in Figure 3.4.6.  This bottomland hardwood stand on the Sherburne Wildlife Management Area in Louisiana was being managed for a generalized (i.e. diversity of taxa) wildlife objective. Photo Credit: Jeremy Stovall

Irregular Shelterwood

The irregular shelterwood method is more of a categorical designation that describes shelterwood systems not adequately reflected in the uniform, group, or strip variants of the system. Typically these systems may be uneven-aged or two-aged, with cohorts varying to a large degree in the age gaps between them. Thus, they are often characterized by an irregular diameter distribution, hence the name of the method. Unlike other applications of the shelterwood, the irregular shelterwood is not an even-aged system. There are three types of irregular shelterwoods that are most common (Raymond et al. 2009).

  • The expanding gap irregular shelterwood is similar in nature to the group shelterwood, and relies on expanding younger cohorts centered on an area of regeneration created by a disturbance. Repeated entries like a selection system are common. In the case of multiple removal cuts, the last cut is called the final cut. Trees removed during the final cut are generally the biggest individuals of the old cohort or individuals likely to rise in value the longer they are kept on the stand (Smith 1986).
  • The continuous cover irregular shelterwood is intended to regenerate and maintain an unbalanced multi-aged stand indefinitely. It works well in situations where there is good growing stock and the objective is maintaining current stand structure (Raymond et al. 2009). In this system, repeated establishment cuts are made, but no removal cuts are performed. It thus is best suited to more shade tolerant species.
  • The extended irregular shelterwood creates a two-cohort stand and is similar in nature to a shelterwood with deferment described in a later chapter.

Ecological Considerations (Silvics)

Shelterwood regeneration creates a moderated microclimate for the establishment of a new cohort provided by the residual trees. It is best-suited for shade tolerant and moderately shade tolerant species, but should be avoided for extremely shade-intolerant pioneer species that will have trouble establishing underneath the older cohort.

When the understory of a stand is too open, the ground will not remain as moist, and will allow grasses and herbaceous species to establish in the open conditions, outcompeting the seedlings. In stands where the understory is too dense, the litter layer can decompose too quickly and make it hard for the seedlings to establish. And even in the cases where they do, a crowded understory can affect the form and survival of the seedling. The goal of the shelterwood method is to open the stand enough to maintain a suitable moisture level on the forest floor for germination and provide the seedlings with an understory that is protected, but allows for them to grow vigorously (Hawley 1946).

Similar to seed-tree regeneration, natural seed dispersal is the primary method of regeneration in these stands. Shelterwood can naturally regenerate some species with heavier seeds because even though some will be predated by wildlife, there are enough trees to produce mast for wildlife to eat and still establish the younger cohort in a high enough density for success. The greater density of seed trees comparable to a seed tree regeneration method will also allow heavier seeds to be more evenly distributed across the entire stand.

The older cohort is often left for three to ten years for most species to ensure adequate regeneration after the establishment cut (Hawley 1946). Trees with good form and genetics that also produce a large quantity of seeds are targeted to be left as the shelterwood trees. While it can be difficult to determine a trees seed-production capacity quantitatively in an operational setting, for many species individuals with a large crown that appears vigorous are good candidates to produce large seed crops. The genetics of the seed trees can determine many characteristics such as growth rate, form, and disease resistance of the newer cohort. As with most silvicultural operations, ‘leave the best and cut the rest’ is a good motto to follow.

Suitable Species

  • US South
    • Loblolly pine (Pinus taeda)
    • Longleaf pine (Pinus palustris)
    • Shortleaf pine (Pinus echinata)
    • Oaks (Quercus spp.)
    • Hickories (Carya spp.)
    • American beech (Fagus grandifolia)
    • Yellow-poplar (Liriodendron tulipifera)
  • US North
    • Eastern white pine (Pinus strobus)
    • Red pine (Pinus resinosa)
    • Eastern hemlock (Tsuga canadensis)
    • Pitch pine (Pinus rigida)
    • Sugar maple (Acer saccharum)
    • Birches (Betula spp.)
  • US West
    • Western white pine (Pinus monticola)
    • Ponderosa pine (Pinus ponderosa)
    • Lodgepole pine (Pinus contorta)
    • Western larch (Larix occidentalis)
    • Douglas-fir (Pseudotsuga menziesii)
    • Western hemlock (Tsuga heterophylla)
    • Sitka spruce (Picea sitchensis)

Economic Considerations

Similar to the seed-tree method, larger trees are not harvested during the establishment cut of the stand and are harvested during a later removal cut. However since there are more trees that are left behind relative to seed tree regeneration methods, the risk of windthrow is lower and more trees are likely to survive in the cases of other low-severity disturbances such as fire, flooding, ice storms, and insect or disease outbreaks. Additional costs will be incurred due to the multiple harvest entries required for most shelterwoods, and capturing value of the overwood will be deferred for a number of years. However these increases in cost are balanced to some extent by the increased value attributed to growth of overwood trees between establishment and removal cuts. One major advantage over the seed tree regeneration method is that there are typically enough trees harvested at the removal cut to make the operation sufficiently attractive to secure a harvest operator.

Compared to a clearcut, a shelterwood requires more skill on a loggers part to avoid damaging overwood trees in the establishment cut, and to avoid damaging regeneration in the removal cut. These are not inherently different from challenges a logger faces in thinning a stand, so it should not be difficult to find a qualified operator in regions of the country with a strong forestry industry. Compared to a clearcut, a shelterwood necessitates more work in marking timber and overseeing each harvest on the foresters part. Sometimes more trees will have to be removed during an establishment cut than originally planned to get a logger to bid on the job. The forester must possess a strong understanding of local economic conditions and available contractors.

One possible outcome of leaving large dominant and codominant overwood trees in the stand following the establishment cut is that some trees may grow larger than the maximum diameter size that local mills can process by the time the removal cut is conducted. It is important to be aware of these conditions when marking the stand for an establishment cut. Additionally, it is necessary to be aware that some trees retained as overwood may just die of various disturbances prior to being harvested. While oversized trees can be left as legacy trees and dead or dying trees can be left in the stand for wildlife purposes, this will depend on the landowner objectives.

In most forest covertypes, if a shelterwood method is implemented properly it can be quite profitable. The low cost of establishment (no planted seedlings or planting costs) coupled with the large number of trees per acre harvested at each entry usually allows each cut in a shelterwood to be a commercial operation.  

Societal Considerations

Shelterwoods tend to be among the most aesthetically pleasing silvicultural systems. Unlike clearcuts, trees are always on the site and unlike the seed tree method, the older cohort is kept on the site as long as possible (Hawley and Smith 1954). In parts of the country with hilly or mountainous terrain, a shelterwood is a far more socially acceptable regeneration method than a clearcut provided the species is suited for it. In fact, in the 1970s, the USDA Forest Service increased the use of the shelterwood method, particularly in National Forests that had previously faced public scrutiny over high-visibility clearcutting (Smith 1986). Additional improvements can be made to the aesthetics of a shelterwood by feathering the boundaries and by creating a gradual transition from the shelterwood site to the surrounding stand. This can produce habitat which provides a shaded environment, conducive to some flora and fauna (Nyland 2002).

Silvicultural System Considerations

Selecting Reserve Trees

The trees best suited to be left as overwood in an establishment cut are

  • members of the dominant or codominant crown classes,
  • that have a high live crown ratios,
  • possess wide, deep crowns, and
  • are characterized by sturdy but narrowing straight boles.

Retained overwood trees are wind resistant individuals that produce a large amount of seeds because they are so vigorous. Short trees with narrow crowns are unsuited to be overwood trees because they are vulnerable to wind damage and may become uprooted (Smith et al. 1997). Relying on damaged trees that produce a large amount of seeds in a last attempt to pass on their genetics is not a good practice for a shelterwood, as those trees may not survive long enough to contribute to the next cohort. Trees that seed prolifically are also good overwood trees because they provide plenty of mast for wildlife, which is a common landowner objective along with timber production.

It is important for the loggers in the stand to be careful not to damage any of the trees that are supposed to remain on the site so that they do not lose quality or become susceptible to infections due to an open wound.

Determining Overwood Density

When making the establishment cut for a shelterwood system, determining the density of the overwood that will be left in the stand is a critical consideration to successfully regenerating the stand. Targeting overwood trees at a specific and uniform spacing should be secondary in marking decisions to factors such as tree quality, composition, and current and potential timber value. However, it is important to cut the stand to a density that will best suit the shade tolerance of the new cohort, while not leaving too much overwood that will suppress growth or shift composition undesirably. Additionally, sufficient volume must be removed to make the operation economically viable. All these factors must be balanced in choosing the appropriate overwood density. The density of the overwood will be primarily dependent on the species that the site is going to be managed for, since some are more shade tolerant than others, and the amount of light penetrating the canopy will vary by species. The table may be used in marking guidelines if target basal areas and mean tree sizes (dbh) for retained overwood can be determined from forest inventory data and an understanding of the silvics of the species in the stand.  

 Table 3.4.1. Overwood densities based on target basal areas by quadratic mean diameter.  BA10 indicates a retained basal area of 10 ft2 / ac. Thus 18.3 trees 10 inches DBH must be left per acre to achieve a retained basal area of 10 ft2 / ac in an establishment cut.

A table listing overwood densities based on target basal areas by quadratic mean diameter.

Establishment Treatments

Site preparation, such as prescribed burning and herbicide, is sometimes used to manage shelterwood sites to limit competition. However, mechanical site preparation involving any sort of tillage should generally be avoided due to the damage it can cause to the existing root systems, retained overwood trees, and the new cohort of seedlings. In the northwest, foresters use shelterwood to regenerate conifers and often use prescribed burning to aid in the disposal of fuels to lower the risk of potential wildfires in the future. Prescribed burning is also used to decrease slash and remove competition. However, foresters should be cautious in order to avoid damaging retained trees of the older cohort. Because of the potential damage establishment treatments can cause, establishment treatments are only used to fix critical site conditions that could become a problem if left untreated (Nyland 2002).  In some cases mulching or herbicide application to remove competing vegetation in the midstory or lower may be necessary.


  • shelterwood3_lockhart

Figure 3.4.9. A uniform shelterwood in a mixed oak stand following the establishment cut in France.  Notice that an exclosure fence has been erected to reduce herbivory, while the site has also been mulched to remove midstory and understory competing vegetation, allowing the new cohort to grow in an adequate light environment. Photo Credit: Brian Lockhart, USDA Forest Service, Bugwood.org

Seedling Mortality During the Removal Cut

It is inevitable when making the removal cut in a shelterwood system that some of the younger cohort will be damaged and killed during the process. While it is best to keep mortality to a minimum by using smaller equipment, the equipment moving over the seedlings can also double as a pre-commercial thinning for the new cohort. If successful, most stands regenerated with a shelterwood will be far denser than required to meet most landowner objectives.  

Intermediate Treatments

In areas where the density is greater than 5,000 trees per acre, there is generally too much competition amongst the new cohort and it will need to be pre-commercially thinned to provide the remaining saplings with enough resources to maintain good form and health. Skidding logs through the stand in the areas that need to be thinned during the removal cut is a fairly easy way to accomplish this at minimal additional cost (Barnett and Baker 1990). Sometimes a precommercial thin will be necessary as the cohort gets older if the density is still too high and affecting growth. One exception to this is where high densities are desirable to force strong apical dominance, reduce forking, and increase self-pruning of branches in hardwood stands with decurrent (i.e. widely branching) crown forms and high-value sawtimber or veneer potential.  

Shelterwood Pros

  • Protection of seedlings from the environment
  • More abundant seed production
  • Locally adapted seed
  • Multiple harvests provide revenue over time
  • Can be used for heavy-seeded species
  • System can be modified to regenerate almost any species
  • Superior aesthetics
  • Can prevent undesirable pioneer species from invading site
  • Due to retention of the best trees, the genetic composition is consistently improving
  • Shorter time between harvests (establishment cuts) because the new cohort is allowed to start growing under the older one
  • Less exposure of mineral soil or potential for erosion
  • Disposal of slash is typically unnecessary
  • Retained overwood trees benefit wildlife

Shelterwood Cons

  • Regenerated cohort could be damaged during removal cuttings
  • Method is more difficult to execute than clearcutting or seed tree
  • With the removal cuttings, seed source and protection is abruptly eliminated
  • Logging is less cost-efficient due to multiple harvest entries
  • Harvesting less valuable trees at the establishment cut makes this a lower-value operation more analogous to a thinning
  • Cannot be used to regenerate extremely shade intolerant species
  • Retained overwood can limit options for establishment treatments



  • shelterwood4_lockhart

Figure 3.4.10. Regeneration following an establishment cut in a bottomland hardwood stand in the US South. Photo Credit: Brian Lockhart, USDA Forest Service, Bugwood.org


  • shelterwood5_lockhart

Figure 3.4.11. The dense new cohort has grown substantially in height following an establishment cut in an upland mixed oak stand in the Ozark National Forest of Arkansas. Photo Credit: Brian Lockhart, USDA Forest Service, Bugwood.org


  • shelterwood6_boyer

Figure 3.4.12. A longleaf pine shelterwood on the Kisatchie National Forest in Louisiana in spring following a 'hot' fall prescribed fire.  The fire was conducted after the establishment cut.  Photo Credit:  William Boyer, USDA Forest Service, Bugwood.org


  • shelterwood7_obrien

Figure 3.4.13. Patchy regeneration following a shelterwood establishment cut in red pine in the US North. Photo Credit: Joseph O'Brien, USDA Forest Service, Bugwood.org


  • shelterwood8_sagor

Figure 3.4.14. A new cohort is establishing in this mixed oak stand following a shelterwood establishment cut in central Minnesota . Photo Credit: Eli Sagor, University of Minnesota, Bugwood.org


  • shelterwood9_roberts

Figure 3.4.15. A shelterwood establishment cut has been implemented on the steep terrain of this Douglas-fir stand in southwestern Oregon. Photo Credit: Scott Roberts,  Mississippi State University, Bugwood.org


Adams, D. L., J. D. Hodges, D. L. Loftis, J. N. Long, R. S. Seymour, and J. A. Helms. 1994. Silviculture Terminology with Appendix of Draft Ecosystem Management Terms. Silviculture Instructors Subgroup of the Silviculture Working Group of the Society of American Foresters. http://oak.snr.missouri.edu/silviculture/silviculture_terminology.htm

Barnett, J. P., & J. B. Baker. 1990. Regeneration methods. Pages 40-41. In M. L. Duryea, & P. M. Dougherty (Eds.), Forest Regeneration Manual (35-50).  Dordrecht, The Netherlands: Kluwer Academic Publishers. ISBN: 07923095596

Hawley, R.C. 1946. The shelterwood method. Pages 117-140. The Practice of Silviculture. 5th edition. John Wiley & Sons, Inc., New York, NY.

Hawley, R.C. and D. M. Smith. 1954. The shelterwood method. Pages 110-142. The Practice of Silviculture. 6th edition. John Wiley & Sons, Inc., New York, NY.

Nyland, R.D. 2002. Shelterwood and seed-tree methods. Pages 313-338. Silviculture Concepts and Applications. 2nd edition. McGraw-Hill Companies, Inc. New York, NY. ISBN: 0073661902

Raymond, P., S. Bédard, V. Roy, C. Larouche, & S. Tremblay. 2009. The irregular shelterwood system: review, classification, and potential application to forests affected by partial disturbances. Journal of Forestry, 107(8), 405-413. http://www.ingentaconnect.com/content/saf/jof/2009/00000107/00000008/art00008

Smith, D.M. 1986. Shelterwood and seed-tree methods. Pages 396-426. The Practice of Silviculture. 8th edition. John Wiley & Sons, Inc., New York, NY. ISBN: 0471800201

Smith, D.M., B. C. Larson, M. J. Kelty, & P. M. S. Ashton. 1997. Double-cohort pure stands regenerated by partial cutting. Pages 347-363. The Practice of Silviculture. 9th edition. John Wiley & Sons, Inc., New York, NY. ISBN: 047110941X

Hits: 5727