By James D. Arney, PhD – Forest Biometrics Research Institute

Fallacy – A mistaken belief, especially one based on unsound argument or reasoning.
Group Selection –

The Dictionary of Forestry published by the Society of American Foresters (SAF) provides the following definition: Trees are removed and new age classes are established in small groups – 1. The width of groups is commonly approximately twice the height of the mature trees with smaller openings providing micro-environments suitable for tolerant regeneration and larger openings providing conditions suitable for more intolerant regeneration. – 2. The management unit or stand in which regeneration, growth and yield are regulated consists of an aggregation of groups.

Executive Summary:

There is no economic or biological advantage which may be assigned to applying a “group selection” regime to a working forest ownership. It is a classroom theory of no practical value. This forest researcher has never found a successful implementation of this concept. The following three pages provide the basis for this statement.


In silviculture and forest planning there are well known and commonly applied major silvicultural regimes. These include clearcutting, seed-tree, shelterwood and selection harvesting approaches. Size of harvest openings, timing of harvest entries and intensity of treatments vary widely within each of these silvicultural systems. Design and implementation of inventory systems and growth models must pay very close attention to the kind and intensity of silviculture anticipated to be applied to any given landscape (ownership). There are two kinds of selection cutting systems commonly identified – Individual-tree selection or Group selection.

However, when the label “group selection” comes up, we must recognize that this is actually a theoretical concept, not a demonstrated practice. It has neither a biological or economic foundation to justify any attempt to implement it in practice. This may be confirmed by using the following very simplistic example.

Group Selection Cutting Regime Example

Assume a 30,000 acre working forest ownership or tract. To simplify the silvicultural relationships, assume flat ground with a common site growth capacity across the entire tract. The most productive (volume and value) rotation age for even-aged regimes has been determined to be 60 years. The entire tract was planted in the recent past. This means that on an annual sustainable basis there will be 500 acres harvested each year and regenerated. To actually implement this management scenario implies that a road network is established where each harvest unit is accessible to log trucks and planting crews.

Assuming flat ground, the impact of a road network across the tract may be minimized. Road engineers are well aware that road networks have higher impacts (acreage and costs) as the topography become increasingly irregular. In this example the entire road is 30-feet wide. Since this is all flat ground the harvest units will be square with the road only occurring along one side of the harvest unit, encroaching one-half (15-feet) of its width into the stands (harvest units) on each side of the road. The roads are straight lines across the entire tract.

Silviculturally, we will apply the group selection treatments in a staggered time series. The objective is to not allow the harvest openings to expand in size as each additional year gets invoked. As expressed in the SAF Dictionary, the management unit may be identified as an aggregation of groups. Figure 1 provides a partial view of stands where each stand (unique color) is divided into five groups. The sequence of harvests will be staggered in time within each stand so that 20 percent of each stand is harvested at each entry. For a 60-year rotation, each stand will experience harvest entries at 12, 24, 36, 48 and 60 years. A minimum 12-year greenup period is provided (by design) between each harvest entry and its neighbors.

Figure 1. Each stand (unique color) is divided into five groups for harvest entries by age.
Computing the Statistics

With this simplistic example in view, it is now appropriate to address the size in acres of each group opening. If each stand aggregate was five acres, then each opening will be one acre in size. The road access will be only on one side of each opening allowing for one-half of the road area within only one side of the opening. This is the minimal impact among all possible scenarios. Table 1 displays the net forested acres and road acres assuming incrementally increasing size of group selection openings with each row. Opening sizes begin at 1.0 acre and end at 60 acres. The table displays the impacts on acreage distributions in each row as though only that size opening was applied across the entire ownership over the full rotation planning horizon (60 years).

Table 1. Distribution of net forested and road acres by size of group selection opening.

The second column from right in Table 1 displays the number of harvest openings per year to maintain this ownership at a sustainable yield capacity through time. There are two major focal points when evaluating the results in this table. These are a) biological yield; and b) economic yield. The first part of economic yield is quite visible. More than six percent additional acres are tied up in roads when applying 1-acre openings than 50-acre openings. Also the entire road network is active every year due to the expansive geographic distribution of openings when applying 1-acre harvest units. No geographic portions of roads or ownership have quiet intervening years when applying 1-acre harvest units. These are significant points of concern for wildlife and watershed management.

Some aspects of biological yield are implied in the SAF definition when referring to shade tolerance of tree species. As is commonly experienced, a number of the higher valued tree species tend to be shade intolerant. This means that these species must be managed using larger opening sizes than shade tolerant species. Otherwise, resulting stocking levels for the target species may be lower or non-existent. In the West this implies that openings must be in excess of one-acre in size. Using Crown Competition Factor[1], trees begin to compete for growing space when open-grown crowns overlap. This crown radius is about 11, 18 or 23 feet for 10, 20 or 30-inch Dbh trees, respectively. Figure 2 displays the proportion of a stand impacted by shade from surrounding mature forest for these three levels of average tree size.

Figure 2. Proportion of stand impacted by shade from surrounding stands.

Typically, foresters consider external influences on growth and yield which create less than a ten percent difference in yield capacity to be insignificant and not operationally feasible to manage. However, differences in growth in excess of ten percent are considered of operational significance and may present opportunities to manage for or against.

In this example it is common to consider the surrounding stands to represent tree sizes in excess of 20-inches in Dbh. This implies that harvest unit openings should be at least 10 acres in size to restrict the shading influence of neighbors to stay at levels of less than 10 percent of this harvest unit opening. For 20-inch Dbh neighbors 16% of a 4-acre opening is affected. For 20-inch Dbh neighbors 23% of a 2-acre opening is affected. These percentage reductions in yield affect the entire 30,000 acres of this ownership when a group selection regime is invoked at these scales.

In conclusion, there is no economic or biological advantage which may be assigned to applying a “group selection” regime to a working forest ownership. It is a classroom theory of no practical value. This forest researcher has never found a successful implementation of this concept.

Clearcutting, Seed-Tree and Shelterwood are even-aged silvicultural regimes. Singe tree selection is an all-aged silvicultural regime. These have all been demonstrated in practice.

[1] Krajicek, J.E., K.A. Brinkman and S.F. Gingrich. 1961. Crown Competition – a measure of density. For.Sci. 7(1):35-42