Welcome to Online Research .....

Most important, effective research paper collection in the world

The effects of transformation of even-aged stands to continuous cover forestry on conifer log quality and wood properties in the UK


  1. Elspeth Macdonald*,
  2. Barry Gardiner and
  3. William Mason
+ Author Affiliations
  1. Forest Research, Northern Research Station, Roslin, Midlothian EH25 9SY, Scotland
  1. *Corresponding author. E-mail: elspeth.macdonald@forestry.gov.uk

Abstract

There is an increasing move in the UK to transform even-aged, single-species conifer plantations to continuous cover forest, i.e. more diverse and irregular stand structures. However, experience of doing this is limited and research to date has not addressed the consequences of this change on timber quality. This paper reviews the impact of transformation on timber quality and wood properties and uses coupled growth and timber property model simulations to examine the effects of different transformation scenarios on Sitka spruce (Picea sitchensis (Bong.) Carr.). The results of the modelling broadly confirmed the conclusions from the literature review. Five key aspects of transformation are considered. Retaining trees to older ages can produce higher quality timber with improved mechanical properties. Regular selective thinning and increased use of crown thinning will improve timber quality, but timing is critical to avoid producing highly tapered trees with heavy branches. Creating gaps in a uniform canopy will generally have a negative impact on the timber quality of the trees around the gap edges. Increased variation in tree age, size, spacing and species will result in greater variation in log quality and wood properties. Using natural regeneration reduces the opportunities for improved growth and timber quality offered by selectively bred planting stock but can deliver good-quality timber if the characteristics of the original stand are suitable and adequate stocking is achieved. The main conclusion of this review is that transformation to continuous cover forestry will not lead to a significant reduction or improvement in the quality of timber being produced in forests in the UK. The main effect will probably be to increase the variation of log sizes and wood properties that are available in the market.

Introduction

This review aims to provide an overview of knowledge about the effects on log quality and wood properties of transforming even-aged conifer plantations to continuous cover forestry (CCF), with particular reference to commercial conifer species growing in the UK. The main focus of the review is the impact of transformation on existing trees, i.e. those in the stand being transformed, although some reference is made to the characteristics of regenerated trees where this is available from the literature. The review is supported by modelling of the impacts of transformation on key wood and tree properties by coupling existing growth and yield models to timber property models.
The forest cover of the UK currently extends to ∼2.8 million hectares, representing almost 12 per cent of the total land area (Forestry Commission, 2008). This represents a major expansion of forestry in the UK from only 5 per cent land cover 100 years ago and was principally the result of the successful establishment of extensive plantations of conifers, generally in upland areas. These plantations are largely composed of stands of single species planted over a short-time period, particularly between 1950 and 1990, resulting in uniform, even-aged forests. The major species are Sitka spruce (Picea sitchensis (Bong.) Carr.), Scots pine (Pinus sylvestris L.), lodgepole pine (Pinus contorta Dougl. ex. Loud.), Douglas-fir (Pseudotsuga menziesii Franco), European larch (Larix deciduas Mill.), Japanese larch (Larix kaempferi (Lamb.) Carr.) and hybrid larch (Larix x europelis Henry).
The majority of plantations are managed using a system of clearfelling coupes of 5–20 ha and replanting with nursery seedlings. Average rotation lengths are generally in the range of 40–60 years, with the result that there has been a steady increase in timber production from UK forests over the last decade, as the stands planted in the second half of the twentieth century mature and are harvested. Production is forecast to continue increasing over the next 10–15 years (Halsall et al., 2006).
Forest policy and public expectations have evolved since these conifer plantations were established and forests are now required to deliver an increasingly diverse range of benefits. The management, conservation and sustainable development of forests were a focus of the United Nations Conference on Environment and Development in Rio de Janeiro in 1992. At a European level, the Ministerial Conference on the Protection of Forest in Europe adopted a number of resolutions relating to the sustainable management of European forests at conferences in Helsinki in 1993 and Lisbon in 1998. There is now a growing emphasis from governments, policymakers and the public on the non-economic benefits that forests can provide, with forest owners and managers being encouraged to provide additional benefits such as wildlife habitats, opportunities for recreation, health promotion, landscape enhancement and carbon sequestration.
These developments led to restructuring of the extensive even-aged plantations in order to reduce the visual impact of large-scale clearfelling (McIntosh, 1995). However, while restructuring of large plantations has created greater diversity ‘between’ the individual stands in plantations (in terms of species, age and size class differences), greater habitat diversity and amenity value are more likely to be achieved by increasing irregularity ‘within’ the stands (Malcolm et al., 2001). The transformation of even-aged stands to CCF (in this paper, CCF is considered to be synonymous with phrases such as ‘Alternatives to Clearfell’ and ‘Low Impact Silvicultural Systems’) is being increasingly used to develop more diverse and varied stand structures. This change in the approach to managing conifer forests in the UK is endorsed in policy documents such as the UK Forestry Standard (Forestry Commission, 2004a) and the audit protocol for the UK Woodland Assurance Standard (UKWAS, 2006) which expects managers to increasingly adopt ‘lower impact silvicultural systems’.
There has been a willingness to adopt CCF where practicable because it generally relies on natural regeneration to establish a successor crop and is seen as a cheaper alternative to the expensive process of ground preparation, planting and tending required with clearfelling–restocking systems (Macdonald and Gardiner, 2005). In addition, some private growers see CCF management as a way of maintaining the capital value of their woodlands, while still obtaining income from thinning operations.
While these more varied approaches to forest management are being adopted, there is limited knowledge and experience of putting them into practice or quantifying the likely outcomes in terms of the impact on timber supply and quality. This is even true of major forestry countries such as Sweden where Agestam et al. (2005) concluded that ‘A general finding was that research and knowledge of managing mixed stands, as compared with monocultures, are limited, which in turn could limit the applicability of mixed stand management’. In the UK, the existing yield and timber quality models for the main commercial conifer species cannot easily be used to predict the effect of these changes on the volume and quality of timber produced. The impact on the supply and quality of timber available to industry for processing is largely uncertain at this stage, although a preliminary study has been made in Wales (Jaakko Pöyry, 2004).
The key elements of the CCF approach being applied to transform conifer plantations in the UK can be summarized (Mason et al., 1999; Pommerening and Murphy, 2004) as
  • management of stands with the objective of increasing species and structural diversity;
  • an avoidance of clearfelling of areas greater than 0.25 ha (approximately two-tree heights wide) without the retention of some mature trees and
  • using natural regeneration rather than planting in most instances.
The application of CCF for conifers in the UK is still at a relatively early stage of development and the available guidance is subject to ongoing review and updating. As a result, the approaches used are guided by broad principles, with managers trying out different techniques to meet their objectives and adapting their management according to the experience gained. It is possible, however, to identify key aspects of CCF management which differ from even-aged clearfell–restock systems in ways that affect tree and stand growth. Therefore, we concentrate on the impact of these factors on log quality and wood properties rather than the impact of specific systems. In particular, the following elements can be expected to be of importance:
  • trees grown to greater ages due to the timescale of transformation;
  • changes to thinning practice;
  • creation of gaps;
  • increased variability in tree size, age and species and
  • limited genetic improvement.

Key log quality characteristics and wood properties

The widely used term ‘timber quality’ can be defined in many ways. One useful definition is ‘all the wood characteristics and properties that affect the value recovery chain and the serviceability of end-products’ (Zhang, 1997). For the purposes of this paper, we will consider some of the key log characteristics and wood properties that affect the ‘fitness for purpose’ of timber for two main end-uses, namely construction and joinery. At the moment, only a very small amount of UK-grown softwood is marketed into joinery end-uses, but this is a higher value end-use that many growers would like to access for a higher financial return.
In construction, the key characteristics of sawn timber are size, mechanical properties and dimensional stability in drying. The mechanical properties of bending stiffness (modulus of elasticity (MOE)) and bending strength (modulus of rupture) are critical for the performance and safety of timber used in building. Mechanical properties of sawn softwood in the UK are normally determined using machine strength grading that bends each piece of timber to assess its MOE or more recently using newly developed X-ray grading technology (Holland and Reynolds, 2005).
For joinery end-uses, the mechanical properties of timber are not usually critical, but the appearance of the wood and its woodworking characteristics (ease of planning, nailing, gluing and ability to take coatings) are important. Appearance grading for joinery takes into account the number and size of knots, whether knots are tight or loose, and the presence of splits, checks, cracks and discolouration (BSI, 1996).
The changes in dimensions that occur as moisture content varies can result in severe distortion of sawn timber. Differential shrinkage of sawn timber during kiln drying can result in significant defects (twist, bow and spring) that may lead to rejection of the material for both construction and joinery.
The tree growth characteristics and wood properties that have a significant impact on softwood timber quality include stem straightness, stem taper, log diameter, growth rate, branching/knottiness, grain angle, wood density, juvenile wood and compression wood. These have been described in detail for Sitka spruce by Macdonald and Hubert (2002) and the most important effects of different growth factors and wood properties on timber quality and performance are summarized in Table 1.
Table 1:
Summary of impact of different tree and wood factors on the wood properties and sawn timber performance of conifers (see Macdonald and Hubert, 2002 for more details)

The effects of transformation on log quality and wood properties

There are very few published studies that deal specifically with the impact of transformation to CCF management on log quality and/or wood properties. In order to evaluate the effects of transformation, we have therefore taken two approaches. The first part of the paper is a review of the published information that is available in relation to the aspects of transformation identified above, i.e. trees grown to greater ages, changes in thinning practice, creation of gaps, increased variability and limited genetic improvement. The second part of the paper uses growth and timber models developed for Sitka spruce (Edwards and Christie, 1981; Gardiner et al., 2005; COFORD, 2008) to explore the impact of a number of silvicultural scenarios where transformation can be expected to influence stand structure.

Review of literature

Trees grown to greater ages and timescale of transformation

The average age of felling in the UK for commercial conifer stands managed under clearfell systems is currently in the range of 40–60 years. A key feature of transformation to CCF is that some trees in the stand will be retained well beyond these standard rotation ages to provide seed and shelter for the regenerating younger crop, as well as for landscape and biodiversity reasons and some trees may be 80–100 years or older.
The result of delaying the felling of the older element of the stand will be taller trees, with larger diameters composed of wood laid down over a longer period, in comparison to those felled at an earlier age. All these factors will have an influence on log characteristics and wood properties. In addition, trees selected as seed bearers are likely to have deep crowns and to be relatively highly tapered (characteristics sought to promote seed production and stability).
Two aspects of retaining trees to older ages need to be considered: first, the impacts on timber quality and second, the operational and marketing implications of harvesting older, bigger trees.

Timber quality

Regular selective thinning, which can be expected to be practised in stands being transformed to continuous cover management, should concentrate increment on final crop trees of good form and maintain an even growth rate. Logs produced from larger diameter, older trees will have a lower proportion of juvenile wood, resulting in a greater volume of mature wood with more desirable wood properties in terms of mechanical performance and drying stability (higher density, lower microfibril angle and lower grain angle).
A recent study investigated the properties of structural timber cut from an 83-year-old stand of Sitka spruce grown in the UK, comparing timber cut from consecutive radial positions within a log (Ridley-Ellis et al., 2008). Both bending strength and stiffness increased substantially with increasing distance from the pith. Timber from the outer part of the logs met the requirements of the C24 strength class in EN338 (CEN, 2003), compared with C14 for timber cut adjacent to the pith. This demonstrates the gains in performance of sawn timber that might be achieved by growing Sitka spruce for 30–40 years beyond current rotation lengths.
Guldin and Fitzpatrick (1991) compared log quality in natural uneven-aged stands of loblolly pine (Pinus taeda L.) with that found in planted and natural even-aged stands. They found that the uneven stands produced logs of a better grade, based on knot sizes, than the even-aged plantations and attributed this difference in part to the fact that trees of a given size were older in the uneven-aged stand. The greatest improvement in log quality was observed in the butt log, cut from the lowest part of the stem, with upper logs from all systems downgraded due to knots.
With older trees, there is the potential to produce a significant amount of knot-free timber in the valuable sawlog part of the stem, once branches have self-pruned, particularly in Scots pine and larch, which lose branches readily. However, there is also the potential to produce extremely large coarse trees with big branches, especially in the case of relatively open-grown Douglas-fir or Sitka spruce. The selection of seed trees with deep vigorous crowns, desirable for stability and seed production, will tend to favour final crop trees with large branches, at least in the upper part of the stem. Similarly, while stem taper is generally lower in older trees, this effect is likely to be mitigated in stands undergoing transformation by the selection of seed trees with high taper. Recent work by Klädtke (2005) has investigated the possibility of maintaining acceptable wood quality in Norway spruce stands by controlling stem taper. He notes that thinning focused on final crop trees will tend to favour more highly tapered trees with low height to diameter ratios to minimize the risk of wind or snow damage but that these trees will also have large branch diameters and wider growth rings. He suggests that maintaining stem diameter at a height to diameter ratio of greater than 60 will result in branch diameters that do not exceed 4 cm, considered the maximum acceptable for good timber quality in the study.

Operational and marketing implications of larger trees

At present mechanized harvesting systems are optimized for the felling and extraction of trees with an average diameter at breast height of 25–35 cm. The limited availability of suitable machinery and trained labour to harvest large conifers may increase the operational costs associated with growing trees to greater ages (Ireland, 2007) but no definitive studies have yet been carried out.
In addition, most modern sawmills in the UK are unable to process logs with a butt diameter greater than 50–55 cm, and so if significant quantities of large diameter logs become available from continuous cover forests, investment in new processing technology will be required. Such investment could be accomplished during the process of transformation, so long as good information is produced on the volumes of large dimension logs likely to become available. Large trees have the potential to supply specific market demands for long lengths and large sections and there is currently a limited established market for these larger dimension sawn products, such as post and beam construction using Douglas-fir. However, many end-uses are now supplied by engineered wood products. Therefore, except in very specific circumstances, there is at present little or no premium on large size logs and generally such material attracts a reduced price (Ireland, 2007).
Poncelet (2004) considers the issue of the quality and marketability of large conifers (Norway spruce, silver fir (Abies alba Mill.) and Douglas-fir) from the forests of Belgium, eastern France, the Black Forest in Germany and Switzerland. He notes that the concept of large conifers has changed with time. Whereas 150 years ago, the French National Forest Inventory defined a large conifer as having a diameter at breast height (d.b.h.) of 46 cm or greater, this has now been reduced to 37.5 cm in line with the processing capacity of modern sawmills. He suggests that there is a danger of large conifers decaying while standing, unless there is an incentive for owners to harvest and sell them into profitable markets. He concludes that poor-quality large trees with big knots and wide growth rings are unlikely to be accepted in high-value markets but that good-quality stems from stands that have been carefully managed have the potential to find good outlets. In order to produce better quality large trees, he recommends either even-aged management with pruning and regular selective thinning up to an age of at least 70 years or management in an uneven-aged, irregular stand where the final crop trees are identified early, favoured in thinnings and retained until at least 80 years of age.
Growing trees to greater ages gives managers an increased opportunity to produce high-quality timber, particularly in the lower part of the stem. To realize this opportunity, the selection of trees to be retained should focus on those with lower taper and finer branching. Although the market for larger logs is limited at present and there is little or no price premium, this could develop as more material of this type becomes available.

Changes to thinning practice

In even-aged conifer plantations in the UK, a standard first thinning generally involves the systematic removal of entire rows of trees to allow access for harvesting machinery, often combined with a selective thinning in the matrix between these racks. Subsequent thinning is generally selective, of the ‘intermediate’ type, involving the removal of most suppressed and subdominant trees together with trees of inferior stem form and also opening up of the canopy by breaking up groups of competing dominant and co-dominant trees. In upland conifer stands, thinning practice is strongly influenced by the risk of wind damage and some stands may be designated as ‘no thin’ owing to their extreme vulnerability to wind-blow (Forestry Commission, 2004b). In addition, thinning of many even-aged plantations may not take place owing to operational constraints or financial cost leading to overstocked stands with smaller trees, many of which are of poor quality.
In continuous cover forest management, and during the transformation process, thinning is the key tool used by managers to manipulate the stand structure, the canopies of potential seed bearing trees and the environment for the establishment of a successor crop. Thinning during transformation will differ considerably from thinning in even-aged stands and will vary according to the silvicultural system employed and the stand structure that is desired. In particular, there will be greater use of crown thinning (Mason and Kerr, 2004) in the early stages of stand development where trees are removed primarily from the upper canopy, i.e. some dominants and co-dominants. The aim of a crown thinning is to give selected dominants or co-dominants freedom to grow rapidly by gradual removal of competing trees. In a shelterwood system, these trees will be those selected as seed-bearing trees to form the overstorey beneath which the successor crop will regenerate, while in a selection system, these trees are likely to form the matrix within which gaps will be opened up around advance regeneration.
Mason and Kerr (2004) provide guidance on thinning strategies for continuous cover stands. Several key elements can be identified:
  • Thinning is likely to be more regular and to continue later in the life of the stand than in even-aged stands;
  • Thinning will favour trees with well-developed crowns and high taper, as these are expected to be most stable (frame trees) and have the greatest capacity to produce heavy cone crops and large quantities of seed (seed trees);
  • Thinning is likely to be slightly heavier than previously practised in even-aged stands in order to improve long-term stability, develop the crowns of seed trees and provide a suitable microclimate for regenerating seedlings. Depending on the timing of first thinning, this could prolong the retention of deep living crowns within the stand;
  • The use of crown thinning will mean that sawlog diameter material will be produced from stands at a younger age than was previously the case because a proportion of dominants will be removed and
  • In selection systems, some of the smaller trees of good form normally removed in an intermediate thinning will be left.
As noted earlier, thinning in stands undergoing transformation to continuous cover systems is likely to be slightly heavier than current standard practice (Mason and Kerr, 2004). For example, Hale (2004) gives guidance on critical values of stand basal area (BA) required for successful seedling regeneration and growth in British conifer forests. Maximum BAs of 20 m2 ha−1 for larch, 25 m2 ha−1 for Scots pine (both light-demanding species), 30 m2 ha−1 for Sitka spruce (light demanding to intermediate), 35 m2 ha−1 for Douglas-fir (intermediate) and 40 m2 ha−1 for western hemlock (shade tolerant) are proposed. Achieving these BAs will require considerably heavier thinning than following the standard thinning regimes assumed in the management tables for even-aged stands, which recommend BAs of greater than 30 m2 ha−1, even immediately after thinning, for most species throughout the rotation (Edwards and Christie, 1981).
Heavier thinning will favour retention of a deeper living crown with vigorous branches. Depending on the time at which a heavy thinning is carried out, this could have a significant impact on log quality resulting in larger knots in the sawlog part of the stem. Heavy thinning is also known to increase stem taper (Brazier, 1977) due to wider spacing. It will also result in increased wind loading on the remaining trees, which tends to lead to increased amounts of compression wood, increased taper (to provide stability) and poor stem form (Jacobs, 1936; Low, 1964; Malcolm and Studholme, 1972).
In order to reduce the impact of a deep living crown, pruning on selected final crop trees can be considered if economically justifiable (Henman, 1963). Pruning limits the extent and size of knots, producing a small knotty core and maximizing the amount of clear timber formed. Pruning also has the effect of advancing crown recession, with the wood in the pruned part of the stem being less under the influence of the live crown (Megraw, 1986). It has been suggested that this accelerates the transition from juvenile to mature wood formation (Briggs and Smith, 1986) and thus reduces the volume of juvenile wood produced. There is also some evidence to suggest that pruning reduces taper (Henman, 1963). Marking and recording of final crop trees is necessary to ensure that the pruned individuals can be appropriately marketed when they are harvested. In Sitka spruce, however, pruning may not result in the production of clear timber. Studies have shown that both heavy thinning and pruning in Sitka spruce can lead to the development of numerous epicormic sprouts (Deal et al., 2003). While these would not affect the suitability of timber for structural purposes (unlikely to be any appreciable loss in mechanical properties caused by the small knots arising from epicormics), the wood produced might not be suitable for appearance-grade applications such as joinery, owing to the frequency of the small knots.
The greater use of crown thinning will affect the quality of timber being produced from the thinning operation and the quality of timber in the remaining crop trees. crown thinning will involve the removal of larger trees at a younger age than is the case during more commonly practised intermediate thinnings. For the grower, this will be likely to be beneficial in financial terms as there will be an increase in the average tree size during thinnings, resulting in lower unit costs for the operation and a higher proportion of sawlogs produced during thinning operations. These sawlogs will be younger for their diameter than logs produced at final felling from no-thin stands or where intermediate thinning has been used. They will tend to have wider growth rings, lower density and a higher proportion of juvenile wood. They can therefore be expected to produce sawn timber with inferior mechanical properties and dimensional stability than equivalent sized but older trees. The trees that remain in the stand after crown thinning will tend to be co-dominants of better form on which, depending on the timing of the thinning, branches on the most valuable part of the stem have already been suppressed. Since they have grown more slowly than the larger trees removed in the thinning, they will have a smaller core of juvenile wood, higher wood density and narrower growth rings. Following the thinning, increment will be added to these trees and they have the potential to grow on to form final crop trees of good form, uniform growth and a small proportion of juvenile wood. Presenting the results of a study on the impact of spacing on timber quality, Brazier (1970) suggested that for Sitka spruce grown at ‘wider spacing’ (i.e. 1.8 and 2.4 m), crown thinning offers an opportunity to produce better quality timber. Its use in the transformation of even-aged plantations to continuous cover systems could be expected, with careful attention to selection of trees and timing of thinning, to significantly improve the timber quality of the final crop. In addition, removal of some of the largest trees during thinning to concentrate growth on slower grown individuals could go some way to alleviate the operational and marketing difficulties associated with the production of large trees grown to greater ages.
Several studies have investigated the impact of different thinning strategies on timber quality. Pape (1999a, b) reported the impact of different thinning treatments on the growth, wood properties and stem quality of Norway spruce growing on highly productive sites in southern Sweden. Different intensities of thinning from below (removing 20 per cent of BA in each of six thinnings, 40 per cent of BA in each of three thinnings or 70 per cent of BA in one thinning) were compared with a no-thin (or ‘natural thinning’ regime) and thinning from above (20 per cent of BA removed in each of six thinnings by removal of the largest trees in the stand). The residual trees in the stands that were thinned from above had a higher average wood density than the unthinned stand and the two heaviest thinning from below treatments. This was due to selection of slower growing trees to form the final crop, rather than to production of lower density wood after thinning. The two heaviest thinning from below treatments resulted in a reduction in wood density compared with the unthinned stand that was attributable to faster growth after thinning treatments were applied. All thinning treatments reduced the juvenile wood proportion compared with the unthinned treatment. This reduction was not any greater for the thinning from above treatment, but the author notes that if trees that have been thinned from above were left to grow on to the same diameter as those thinned from below, then a greater reduction in juvenile wood content could be expected. Crown characteristics were also studied. The only difference in branch diameter (assessed as the diameter of the thickest branch between 1 and 2 m above the ground) was between thinning from above and the heaviest thinning from below treatment. This difference was attributed to selection as the live crown had already receded above 2 m before thinning was carried out. As would be expected, the height to the first live branch was greatest for the unthinned stand (showing greatest crown recession) and lowest for the most heavily thinned from below stands. Stem taper between 1.3 and 4 m up the stem was lowest for the unthinned stand and highest following the heaviest thinning from below. The author concluded that thinning from above can have a positive impact on timber quality but notes that the risk of wind or snow damage could be increased. He also suggested that heavy thinning from below on a reduced number of occasions, removing up to 40 per cent of BA, would not substantially affect wood properties.
Seeling (2001) examined log quality, wood properties and sawn timber performance in Norway spruce from a thinning trial that included heavy thinning treatments that were similar to those that might be applied during transformation of an even-aged stand. The first thinning treatments were applied at age 26 when the trees were growing at a stand density of ∼5000 stems ha−1. The timber was examined 23 years later following a series of thinnings that had reduced stand density to between 236 and 645 stems ha−1.
When roundwood quality was assessed, it was found that the widest spacing after thinning was associated with increased mean ring width, higher taper, bigger branch stubs, increased spiral grain, more compression wood and greater pith eccentricity in log ends. These differences represented a reduction in the quality of timber from plots that had been most heavily thinned, which was reflected when logs were graded according to the European Standard (1996) EN 1927-1 (Figure 1).
Figure 1.
Roundwood quality (volume weighted) according to EN 1927-1 (D, worst quality), from Seeling (2001).
Examination of the sawn timber showed increases in maximum knot diameters and amount of compression wood in material from the widest spaced plots, but these were not statistically significant. After kiln drying, increased twist was found in timber from the widest spaced plots, which was thought to be associated with the increased amounts of compression wood.
Jähagen and Lageson (1996) compared the effects of thinning from above (i.e. crown thinning) and thinning from below on the timber quality of the residual stand after a first thinning in Scots pine. They assessed the diameter of the thickest branch, stem taper, ring width, stem straightness and stem lean. The only significant differences that they found were for stem lean and stem straightness, where thinning from above resulted in fewer leaning or crooked stems. They suggested that the lack of differences in other characteristics may be due to the fact that in a first thinning, all the trees in access racks for extraction have to be felled, reducing the element of selection in the thinning.
In summary, changes in thinning practice will be an integral part of the transformation of even-aged stands to CCF. Some aspects, such as the regularity of selective thinning and greater use of crown thinning, have the potential to improve the quality of timber produced. The timing of thinning operations is critical. Heavy thinning of younger stands could produce trees with increased taper, larger branches and an overall deterioration in timber quality.

Creation of gaps

In even-aged management systems, trees are planted at a uniform spacing and subsequent thinning is normally designed to maintain this uniformity of stocking and avoid opening up any large gaps. However, transformation to continuous cover can, in some situations, involve the creation of gaps in the canopy that will have an impact on the remaining trees.
In the uniform shelterwood system, the canopy is opened up evenly across the stand with each tree receiving a similar increase in light and wind loading. In other systems, gaps are created around advance regeneration or to allow sufficient light in to promote regeneration. These larger gaps create gradients in the light environment and variability in wind loading.
In some silvicultural systems (e.g. group selection), there are a greater number of gaps created than in even-aged management. Some of the effects of gap creation will be similar to those described above for heavier thinning, e.g. retention of deep living crown, and increased growth rate. The particular aspects of gap creation that may have a different impact on timber quality to heavy thinning are the creation of an uneven light environment across the canopy of the edge trees and an increase in wind loading.
Watson and Cameron (1995) studied the effects of growing Sitka spruce in a nursing mixture with Japanese larch (L. kaempferi (Lamb.) Carr.) or lodgepole pine. They found that after 25–30 years, the differential growth rates, where the spruce had outgrown the nurse species, had produced spruce with large uneven crowns, resulting in greater knot areas and increased compression wood formation. It is probable that the creation of gaps in group shelterwood or group selection systems will produce similarly unbalanced crowns on the trees on the edge of the gap. Depending on the age at which the canopy is opened to form the gaps, extremely large knots could form in the sawlog producing part of the stem as branch growth into the gap increases in response to the additional light. This is similar to growing the trees at very wide spacing, which has been shown to produce very large branching and knots (e.g. Brazier and Mobbs, 1993).
Creation of gaps will result in the production of a greater number of ‘edge’ trees than is normally the case in even-aged plantations. There will be increased wind loading on these trees (Gardiner et al., 1997). The effects of increased wind loading can include the development of leaning stems with elliptical cross-section, compression wood formation, increased taper, increased grain angle and poor stem straightness (Jacobs, 1936; Low, 1964; Malcolm and Studholme, 1972; Nicholls, 1982). All these effects are likely to have a negative impact on timber quality compared with stands where a uniform canopy density is maintained and only opened up gradually (Brüchert, 2000).
Quine (2004) studied the development of epicormic sprouts on Sitka spruce stems following the formation of wind-thrown gaps. He found that the production of epicormic sprouts was widespread in Sitka spruce and that larger gaps and road-line felling resulted in longer, thicker epicormic branches that survived for longer than those on trees in smaller gaps. These results suggest that the opening up of gaps during the transformation of Sitka spruce stands is likely to promote the development of abundant epicormic sprouts on the stems of trees around the edge of the gap. As described above, in relation to heavy thinning and pruning, the practical impact of epicormic sprouting is unlikely to be significant unless a joinery market is sought.
The creation of gaps in a uniform canopy is likely to have a generally detrimental effect on the timber quality of trees around the edge of the gaps. The practical importance of this impact will depend on the age of the trees when gaps are created, the species and the intended end-use for the timber.

Increased variability in tree size, age and species

A major objective of thinning in even-aged stands is often to enhance uniformity, whereas in continuous cover, the aim can be to enhance variability. This increased variability during the transformation phase is likely to be in terms of spacing between trees, tree heights and diameter distributions and crown characteristics. This will lead, everything else being equal, to greater variation in timber properties. As the successor crop begins to regenerate, there can also be an increase in the variability of species and age class. This increase in variability will be greatest in selection systems and least in shelterwood systems. Such variability can be measured using structural indices such as those developed by Pommerening (2002).
In selection systems, in particular, there will tend to be increased variability in tree growth characteristics and consequent timber quality. For example, a review of the timber quality of mixed spruce and beech stands in Germany by Knocke and Seifert (2008) suggested that there was a decline of quality along edges between blocks of the two species with an increase in crown variability and asymmetry. However, this was compensated by an increase in stand resistance to damage. During transformation, variable spacing and the creation of gaps will have some of the impacts described in the previous sections relating to growing trees to greater ages, thinning and gap creation. The move away from more uniform growing conditions will result in some trees having increased taper, less even crown development and greater compression wood development. At the same time, careful selective thinning and growing older trees could also produce some high-quality stems with superior wood properties. In addition, particularly in selection systems, there will be a greater range of tree sizes in stands at any one time.
The variability in characteristics and size of timber being produced during transformation to CCF will necessitate a greater degree of pre-sorting in the forest and development of innovative techniques for assessing timber quality prior to or during harvest. This could ensure that timber is targeted to the end-use for which it is best suited and optimize value recovery for both growers and processors.

Limited genetic improvement

As described previously, natural regeneration is generally the preferred method of establishing a successor crop in CCF systems. Using natural regeneration reduces the costs associated with planting, particularly where it is used to establish or maintain mixed, uneven-aged stands for which planting costs are generally higher as a result of the scattered, smaller areas being regenerated. With natural regeneration, however, the species and genetic make-up of the successor crop will be consequent on the stand from which it is being regenerated. The only opportunity for improvement in terms of vigour, stem form or other characteristics will be through the use of selective thinning to favour better trees. With successful regeneration, there are likely to be some benefits of high initial stocking because of the possibilities of choosing the best phenotypes during the initial ‘cleaning’ process and early thinning but with the additional costs of this process.
The use of natural regeneration in transformation to continuous cover systems reduces the opportunity to make improvements in terms of growth and wood properties through the use of genetically improved plants. The Sitka spruce breeding programme in Great Britain has produced significant genetic gains in terms of growth and stem form, without a reduction in wood density when compared with unimproved material (Lee, 1999; Lee, 2004, Mochan et al., 2008). Seed lots are commercially available from different improved production populations, allowing growers to choose, for example, between improved growth and stem form without a reduction in wood density or improved density with a smaller increase in growth and similar stem form improvement. For Scots pine, seed orchard material can also provide improvements in diameter growth and straightness (Lee, 2004). For other species where there is less opportunity to use improved material, the impact of using natural regeneration may be less direct. However, if the form and quality of the existing crop trees are poor, then the regenerating trees may also be poor and replanting or enrichment with better material might be the only way to improve the situation.
In stands that are of good quality in terms of stem form, branching characteristics and growth, the use of natural regeneration may have some advantages compared with planting. Successful natural regeneration tends to result in much more densely stocked stands than planting, which are normally re-spaced at ∼3 m in height. The increased stand density results in early suppression of branches in the lower stem, lower juvenile core area, improved stem straightness and allows a greater element of selectivity in thinnings as there are a larger number of trees to choose from. For example, Klang and Ekö (1999) found improvements in Norway spruce established with the shelter of birch because of a reduction in branch diameters and the incidence of ‘spike-knots’ when compared with spruce established with no shelter. Auty and Achim (2008) found that the mechanical properties of naturally regenerated Scots pine, predicted from an assessment of acoustic velocity in standing trees, was at least comparable with and perhaps slightly better than plantation-grown material of a similar age.
There is also some evidence that young trees in stands grown from direct seeding develop better formed root systems than planted ones (Pfeifer, 1982; Watson and Tombleson, 2002). This can reduce the incidence of toppling as a result of wind or snow damage in the young trees and may also have an impact on stem form and longer term stand stability. Naturally regenerated trees could be expected to exhibit similar characteristics to those grown from direct seeding but the full benefits need to be quantified.
Careful management of successful natural regeneration, where the parent stand is of good form and well suited to the site, has the potential to produce good-quality timber from stable trees established at a high stocking density and selectively re-spaced and thinned. Where inadequate stocking is achieved, or the existing crop is of poor form, supplementary planting may be required and the opportunity can be taken to achieve timber quality improvements by the use of selectively bred planting stock.

Summary of review of literature

Growing some trees to greater ages, which is likely to be a feature of many stands undergoing transformation to CCF, offers the potential for the production of higher quality timber with improved mechanical properties. The economic return will depend on the availability of harvesting and processing equipment able to tackle larger trees and the development of markets for such material.
The regular selective thinning and increased use of crown thinning that are likely to take place as stands are transformed to CCF could deliver improved timber quality, providing operations are gradual and timed carefully to avoid opening up the canopy too much in younger stands, resulting in increased taper and branch size.
Where gaps are created in even-aged stands, there is likely to be an overall reduction in timber quality of the trees around the edges of gaps, which may develop imbalanced crowns with some very large knots and an increased incidence of compression wood.
Where there is increased variation in tree age, size, spacing and species in stands undergoing transformation, greater variation in log quality and wood properties can be expected. Optimizing the utilization of this timber and maximizing its value will require the use of innovative timber quality assessment techniques.
The use of natural regeneration to establish successor stands during transformation reduces the opportunities for realizing gains in growth and timber quality offered by selectively bred planting stock. However, where the existing stand is of good form and adequate stocking is achieved by natural regeneration, followed by selective re-spacing and thinning, good-quality timber may be produced.

Modelling the impacts of transformation

In order to further explore the potential impacts of transformation to continuous cover on log quality and wood properties, we coupled growth and timber property models to predict changes that could be anticipated under different silvicultural scenarios. Currently for British grown conifers, there are only timber property models for Sitka spruce and all the modelling work describes the expected impact of transformation on Sitka spruce. Since the majority of stands undergoing transformation will be Sitka spruce, this is felt to be an acceptable first step.

Description of models

The yield models of Edwards and Christie (1981) (hereafter referred to as E&C) and National Council for Forest Research and Development in the Republic of Ireland (COFORD, 2008) for Sitka spruce were linked with the Sitka spruce timber quality model of Gardiner et al. (2005) to simulate some of the possible wood property changes that might be expected under different simulations of transformation from even-aged to CCF. The E&C models are the standard for Sitka spruce growing in Britain and act as the reference in this paper. Unfortunately, they only describe standard management practice and, therefore, the COFORD model was used to simulate transformation. Comparison between the two models was carried out to show how well the COFORD model agreed with the E&C model for normal thinnings. It also needs to be remembered that these growth models were developed for even-aged Sitka spruce and the simulations can only be regarded as approximations to the way spruce will grow under the different silvicultural treatments.
The output of the yield models provides the stand mean d.b.h. and height for each year up to the age of the simulation, which are needed as inputs to run the timber property models. The tree height and diameter estimates are used as inputs into a taper function which specifies an annual average stem profile for the mean tree in the stand. A tree density profile is then derived based on calculated ring widths from the taper model and ring number (from the pith) for all locations in the tree stem. The taper and density models are described in detail in Gardiner et al. (2005). Other properties important for timber performance are also derived within the stem, including knot size and number (Achim et al., 2006) and grain angle (Mavrou, 2007).

Modelling approach

The approach has been to simulate a number of representative scenarios at two different productivity classes, as represented by Yield Class (YC) (Edwards and Christie, 1981), and a number of ages and to compare the results between scenarios. Only the mean tree for each scenario has been modelled because we have insufficient information on diameter and height distributions in the different silvicultural systems to reliably look at tree-to-tree variation.
The timber quality model outputs have been chosen to provide information about volume availability and recovery and likely timber performance. The information is provided for the whole tree and also for the first 10 m because most of the sawn timber is obtained from the lower part of the tree. Specifically, the model outputs are
  • • Volume and recovery information
    • ○ mean tree size (height and d.b.h.);
    • ○ volume (to 16 and 7 cm over-bark) and
    • ○ taper (whole tree and first 10 m).
  • • Timber performance (strength and distortion)
    • ○ wood density (minimum, mean for tree, mean in first 10 m);
    • ○ height of the first live branch;
    • ○ size of the largest branch on the outside of the tree;
    • ○ knot area ratio on the outside of the stem in first 10 m;
    • ○ mean ring width;
    • ○ ratio of largest ring width to mean ring width; this provides a measure of the evenness of growth across the tree and
    • ○ grain angle at the outside of the tree at a height of 1.3 m.

Scenarios

Four scenarios were developed to simulate a range of management approaches that might be adopted to transform even-aged stands to continuous cover forest management. These scenarios were compared with a ‘control’ scenario of a standard clearfell and restock even-aged management system. The tested scenarios are shown in Table 2 and were
1 Standard management
Table 2:
Predictions of log and wood properties for mean trees under different simulated silvicultural systems, different ages and different YCs (thinning type: IZ = intermediate thinning; CZ = crown thinning; HT = heavy thinning)
This scenario was designed to simulate standard even-aged Sitka spruce silviculture, using a rotation length close to the average for Britain. Four different model simulations were run:
  • a  Intermediate thinning at 5-year intervals using E&C yield models. Initial spacing—2.0 m; YC—14; age at felling—45 years. This scenario acted as the control for all subsequent E&C simulations, using an average productivity for Sitka spruce in the UK.
  • b As for 1a but YC20, representing a high-productivity site.
  • c  Intermediate thinning using COFORD yield models. Model initialization (starting age, d.b.h., height and BA) and BA removal at 5-year intervals as in E&C yield models (1a above). Initial spacing—2.0 m; YC—14; age at felling—45 years. This simulation was used to check whether there were any differences between the predictions of E&C and COFORD models, for the same scenario. It also acts as the control for subsequent COFORD simulations.
  • d As for 1c but YC20.
2 Irregular shelterwood
In this scenario, trees were grown for longer to simulate the impact of conversion to an irregular shelterwood, with some trees retained for much longer than normal. The trees for which the model simulation was run were those retained to older age.
  • a Intermediate thinning at 5-year intervals using E&C yield models. Initial spacing—2.0 m; YC—14; age at felling—80 years.
  • b As for 2a but YC20.
3 Crown thinning
This scenario was used to investigate the impact of increased use of crown thinning, comparing the log characteristics and wood properties of the final crop trees following crown thinning with those of final crop trees following intermediate thinning.
Unfortunately, there are no E&C crown thinning models available with an initial spacing of 2.0 m. Therefore, the crown thinning model runs use an initial spacing of 1.7 m and for direct comparison, an intermediate thinning with an initial spacing of 1.7 m was also run.
  • a  Crown thinning at 5-year intervals using yield models of E&C. Initial spacing—1.7 m (only spacing available); YC—14; age at felling—45 years.
  • b As for 3a but YC20.
  • c Intermediate thinning at 5-year intervals using yield models of E&C. Initial spacing—1.7 m; YC—14; age at felling—45 years.
  • d As for 3c but YC20.
4 Frame-tree thinning
Mason and Kerr (2004) describe a system for converting even-aged stands to a more complex structure by identifying 40–80 ‘frame’ trees at the start of the transformation process and favouring these in all subsequent thinnings. Our model simulation is designed to look at the properties of the matrix trees in between the frame trees and not the frame trees themselves. The model simulations were compared with actual measurements of a frame-tree thinning in Coed Trallwm, Wales (Price, 2007), to ensure they were realistic.
  • a  Using COFORD yield models. Initialized by E&C yield models followed by two intermediate thinnings and by two crown thinnings, all at 5-year intervals. BA removal for the thinnings as per E&C models. Initial spacing—2.0 m; YC—14; age at felling—45 years.
  • b As for 4a but YC20.
5 Group felling
In this scenario, we simulate the opening up of a gap in an even-aged, single storey canopy to allow natural regeneration to occur. Only the log characteristics and wood properties of the trees around the edge of the gap are modelled.
  • a  Using COFORD models. Initialized with E&C yield models, followed by one intermediate and one crown thinning at 5-year intervals (BA removal as per E&C models) and a very heavy neutral thinning (∼70 per cent BA removed to leave 10 m2 ha−1) after a further 10 years. Initial spacing—2.0 m; YC—14; age at felling—45 years.
  • b As for 5a but age at felling 60 years (maximum possible with COFORD model).
  • c As for 5a but YC20.
  • d As for 5a but YC20 and age at felling 60 years.

Model simulation results

Model outputs are presented in Table 2. The trees modelled are the trees left in the stand at the felling age shown, not those removed during thinnings. In the following discussion, intermediate thinning at YC 14 and an initial spacing of 2.0 m (both E&C and COFORD) will be regarded as the control against which all the other simulations will be compared.

Effect of growth rate

An increase in YC from 14 to 20 using the E&C models (1a vs 1b) produces an increase in tree size, log volume and a decrease in juvenile wood percentage but there is no impact on taper. Average and minimum wood density both decrease. The knot area ratio on the outside of the tree is reduced with increasing growth rate but the largest whorl branch is bigger. Mean ring width is larger as expected but the ratio of maximum ring width to mean ring width is lower suggesting more even growth on average at these higher growth rates. More even growth rate will give lower variation in the wood properties of sawn timber leading to better mechanical performance and less distortion.
The results from the simulations using the COFORD model (1c and 1d) are very similar except for the wood density results. These show less difference in the average density between YC 14 and 20 but a marked reduction in the minimum density. This appears to be due to the fact that the COFORD model gives annual growth (E&C model annual growth is interpolated from 5-year values) and the density model responds to the sudden increase in growth rate following thinning (see Figure 2).
Figure 2.
(a) Simulation of the density (kg m−3) distribution through stem of 45-year-old Sitka spruce growing at YC14 with a 2.0 m initial spacing and managed with intermediate thinning. (Tree growth simulated using COFORD model). (b) Density slice at 1.3-m height from simulation displayed in Figure 2a.

Effect of age at felling

Timber properties at different ages (45 and 80 years) were compared using the E&C intermediate thinning models (1a vs 2a; 1b vs 2b). Apart from the obvious increase in tree size and log volume, there is a very large reduction in juvenile wood percentage (approximately halved) because there is a much larger increment of mature wood. This additional mature wood also leads to an increase in the average wood density and a reduction in the grain angle at the outside of the tree. The knot area ratio in the bottom part of the tree is reduced and the relative height of the base of the live canopy is much higher on the tree as the canopy lifts with age. However, the mean ring width is lower overall because the radial growth slows giving on average less even growth (maximum ring width/mean ring width).

Effect of increased use of crown thinning

A comparison of crown and intermediate thinning was made using E&C yield models (3a vs 3c; 3b vs 3d). The sizes of the trees and logs under crown thinning are lower, as is expected. To get trees and logs of similar size to an intermediate thinning, it would be necessary to grow the trees for an additional 3–4 years. There is a reduction in the juvenile wood percentage as expected (although weak for YC20) and an increase in height to diameter ratio (less tapered) following the crown thinning. The average wood density is also higher and the knot area ratio is reduced but the evenness of growth is slightly reduced. Overall crown thinning produces improved wood properties but the retention period would need to be longer to obtain the same size of trees. At the same time, it needs to be remembered that the trees removed during crown thinning will tend to have larger juvenile cores than thinnings from more conventional management.

Implication of frame-tree thinning

In this scenario, the matrix trees following the simulated frame-tree thinning are compared with the matrix trees following intermediate thinning using the COFORD yield models (1c vs 4a; 1d vs 4b). We do not look at the properties of the frame trees themselves, which will be considerably bigger. Following a programme of frame-tree thinning, the matrix trees have a smaller d.b.h. and volume on average compared with intermediate thinning and the percentage of juvenile wood in the stems is higher. The mean wood densities are higher both in the lower part of the tree and in the tree as a whole despite the higher percentage of juvenile wood. This is because the frame-tree thinning simulation uses intermediate thinning followed by crown thinning, so that average growth rates and ring widths are lower and there is a corresponding increase in average wood density. Knot area ratios and branch size are smaller and the crown is at a higher relative height. In general because the matrix trees are smaller than would be the case in an intermediate thinning, the wood properties are improved (high wood density and less knotty) but the problem is that the trees will contain a high percentage of juvenile wood with poorer properties (high MFA, low stiffness and increased distortion).

Implication of group felling

Simulated group felling was compared with intermediate thinning using the COFORD yield models (1c vs 5a and 5b; 1d vs 5c and 5d). This is not a true group felling and it was only possible to mimic opening a gap in the canopy by a very heavy thinning. This means there is no eccentricity in the tree characteristics and the characteristics are relevant only for the side of the tree facing a gap. Tree and log sizes are much larger, the juvenile wood percentage is reduced and the height to diameter ratio is decreased (stems more tapered). Minimum and mean wood densities are slightly reduced in both the lower part of the tree and the whole tree at YC14 but slightly increased at YC20. The height of the live crown is proportionally lower in the group felling, while the largest branch, knot area ratio in lowest 10 m and mean ring width at 5 m are larger. With increased age (5b and 5d), the juvenile wood percentage decreases further, the average wood density increases both for the whole tree and the lowest 10 m, the crown lifts further off the ground and the knot area ratio decreases compared with these values at 45 years (5a and 5c). The largest branch and the mean ring widths are still larger than for any other simulation. However, the detrimental effect of larger branch sizes will be more in evidence in the upper part of the tree than in the lower 10 m, which produces most of the sawn timber.
Generally, the edge trees in a group felling system will have poorer wood properties and timber performance than timber from other management systems. The trees within the matrix (away from the edge) will tend to have properties closer to the matrix trees in a frame-tree thinning (4a and 4b).

Summary of modelling

It was not possible to model the exact details of alternative silvicultural systems with the yield models currently available so the results above have to be treated with caution. However, the general conclusions are very similar to those in the first part of the paper based on a review of the literature. The longer growth periods more typical of CCF are likely to produce higher quality mature wood with less juvenile wood, higher wood density and lower knottiness. In addition, the changes in thinning practice that give more emphasis to crown thinning (including frame-tree thinning) have the potential to produce high-quality wood, although there will be an increase in large logs from these thinnings, which have an extensive juvenile core. On the other hand, group felling will lead to a larger variability of tree and wood properties within stands with some trees on the edges of gaps having much poorer characteristics and form. In all cases, careful selection of trees for thinning and those designated as final crop trees will be crucial.
The model results presented were for Sitka spruce alone because currently this is the only species for which timber property models exist that are applicable under UK conditions. With other species, the results are likely to be somewhat different. For example, there will be very different growth responses from trees that are more shade tolerant (e.g. western hemlock) or less shade tolerant (e.g. Scots and Corsican pine and larch) than Sitka spruce. With the silvicultural systems described above, shade-tolerant species will tend to show less variability in growth rate compared with Sitka spruce whereas shade-intolerant species will tend to show more variation. In addition, changes in wood properties resulting from different silvicultural systems will vary from species to species. For example, the density of the hard pines like Scots pine and Corsican pine as well as larch will be less affected by changes in growth rate when compared with Sitka spruce or Douglas-fir (Macdonald and Hubert, 2002). Furthermore, changes in branch size will be more important in the pines than spruce or Douglas-fir because knot size is more often a limiting factor in assessing timber quality in pines. In summary, the modelling work described is preliminary, limited in applicability to Sitka spruce and also requires further validation against field measurements. To fully assess the implications of transformation to CCF in British forestry, there needs to be continued development of timber property models for the key UK forestry species and these need to be linked to new growth and yield models capable of fully simulating CCF systems under UK conditions.

Conclusions and recommendations

The use of CCF in conifer forests in the UK is increasing. However, there is limited experience of implementing this approach on a large scale, and it will take time to build up expertise and knowledge in this area. In recent years, most of the research and development in this subject has been focused on stand manipulation to achieve successful natural regeneration, with little emphasis on the quality of timber that will be produced from these stands.
From the available information, it can be concluded that CCF has the potential to produce high-quality timber, providing the quality of the original stand is of a sufficient standard. A key factor is likely to be continuity of management, as the transformation process may take up to 50 years and subsequent continuous cover management will require an ongoing commitment to monitoring and control. Careful silviculture with particular attention to the timing of interventions in relation to crown development and to selection during thinning is required. There is likely to be greater variation in the characteristics of future timber supplies in terms of size and wood properties. Wider use of crown thinning will result in an increased quantity of young sawlog-size material being processed in sawmills, which is likely to have inferior mechanical properties and dimensional stability than older material of the same dimensions. The exact volumes of this young sawlog-size material and the practical impact of these changes in terms of timber performance and value have yet to be evaluated.
The impact of these changes in stand management on timber quality and supply at a national level is hard to judge without a realistic estimate of the location, extent and characteristics of the forests being transformed. Impacts will be greatest in stands being transformed to or being managed using selection systems. Mason (2003) states that a preliminary site evaluation suggests that perhaps 50 per cent of sites in Wales and 30 per cent of sites in Scotland might be suitable for transformation. In order to better gauge the effects of increased use of continuous cover systems, a systematic inventory of the area and type of silvicultural system planned for forests in the UK will be required.
Our ability to predict the impact of alternative silvicultural strategies using currently available growth and timber property models is limited, as these have been largely developed for use in even-aged, single-species stands. There is a need to continue to adapt and improve models for the main commercial conifer species in the UK (Sitka spruce, Scots pine, Douglas-fir and larch), so that they can deal with the range of silvicultural systems that are being used. The establishment of long-term experiments to test these models and provide research material for the future would also be an important step forward.
On the basis of information that is currently available, the following recommendations can be made, with the aim of ensuring that the quality of timber produced from stands undergoing transformation is as good as possible:
  • 1  As far as possible, opening up of the canopy beyond the extent of a conventional intermediate thinning should be avoided until the live crown has receded to a height that is above the most valuable sawlog-producing part of the stem.
  • 2  In all thinnings, trees with straight stems, superior branching habit (small branch diameters, few branches per whorl and level branch insertion angle) and low taper should be favoured.
  • 3  Crown thinning should be used carefully to favour the best formed co-dominants in order to produce final crop trees with a low proportion of juvenile wood.
  • 4  Where the canopy has to be opened up more radically to release natural regeneration, and where the seed trees selected are highly tapered with vigorous crowns, pruning should be considered to improve the quality of the final crop trees if economically justifiable.
  • 5  If the existing crop has inferior stem straightness and branching, or is not well suited to the site, planting to introduce improved genotypes or provenances or preferred species should be used to supplement natural regeneration.
  • 6  Where natural regeneration is successful in producing a high stocking density, selective re-spacing and thinning should be used to identify the final crop trees.
  • 7  In order to realize the potential for CCF management to improve the quality of coniferous timber produced in the UK, regular monitoring and thinning are essential.
Forestry in the UK is rapidly evolving to meet the increasing demands being placed upon it. CCF is an approach that can deliver additional social, environmental and economic benefits. However, little is known about the consequences of this transformation on the future timber supply available from UK forests. As this review has shown there is a lack of information on the impacts of transformation on either the future wood quality or timber volume, nor are there comparisons of the differences between different systems. For this reason, many of the points and conclusions reached in this paper are of a general nature and it is difficult to give exact advice and recommendations. Therefore, it is important as soon as possible to
  • 1  Continue to develop growth and yield models that adequately represent the growth of trees in the key silvicultural systems being incorporated under CCF within UK forestry management.
  • 2  Make a full assessment of the likely areas being transformed together with the systems being used and the timescale for conversion. This can then be coupled with the models developed above to predict the volume and size assortments of timber from UK forests over the next 20–50 years. This is a key requirement for investment in the forestry/wood-processing sector.
  • 3  Conduct full-scale sawmill trials comparing the quality and volume of timber from different silvicultural systems used in CCF against a uniform clearfell–restocking control, if sites representing such comparisons are available in the UK.

Funding

Forestry Commission and the Scottish Forestry Trust.

Conflict of Interest Statement

None declared.

Acknowledgments

Thanks are due to Gary Kerr for helpful comments on drafts of this paper and to Forest Research Library staff for their assistance with sourcing publications. The authors are grateful to James Guldin and an anonymous reviewer for their valuable suggestions for improving the paper which were greatly appreciated.
The online version of this article has been published under an open access model. Users are entitled to use, reproduce, disseminate, or display the open access version of this article for non-commercial purposes provided that: the original authorship is properly and fully attributed; the Journal and Oxford University Press are attributed as the original place of publication with the correct citation details given; if an article is subsequently reproduced or disseminated not in its entirety but only in part or as a derivative work this must be clearly indicated. For commercial re-use, please contact journals.permissions@oxfordjournals.org

References

কোন মন্তব্য নেই:

একটি মন্তব্য পোস্ট করুন