1

Re-evaluating, understanding and selecting Welsh softwoods; using wood variability to create value

Dainis Dauksta February 2015

2

Executive summary 1. Sheltered areas of Wales are amongst the best places in Britain for growing conifers.

2. Long term retention of conifer stands on sheltered sites will produce higher grade timber

and as self-pruning occurs will also create higher amenity value.

3. Large diameter homegrown conifers can yield high grade, high value large dimension

softwoods not easily available from Scandinavian or Baltic forests.

4. It is incorrect that fast growth of conifer species in Britain produces low grade timber.

5. Consistency of growth rate is more important than growth rate considered alone for

production of high grade timber.

6. Wood qualities need to be defined in terms of end use and fitness for purpose otherwise

wood quality becomes an arbitrary term.

7. Ideologically motivated commentators have negatively influenced political and public

perceptions of planted forests.

8. Planted forests are now considered to be foundations of sustainable economies by WWF &

FAO but are seen by FAO as subject to socio-economic risk due to governance failures.

9. Planted forests make up only 7% of world forest cover but supply around half of global

industrial roundwood demand and sequester 1.5 gigatons (1.5 x 109 tons) of carbon p.a.

10. According to FAO, trends in European forest management could result in a shortfall of

coniferous timber; according to WWF this will happen as world demand for timber

dramatically increases.

11. Scottish and Welsh researchers have demonstrated that British softwoods can be used in a

range of contemporary construction techniques including mid-rise multi-storey structures.

12. Wales has sites ideal for growing valuable high grade softwood species such as Douglas fir;

this timber can be sold for the same price as oak. Western hemlock makes an ideal

understorey, it is also a useful joinery softwood. Larch has been underutilised for many

years, with its increased availability we now have opportunities to re-evaluate it.

13. Warm rainy Welsh autumns promote growth of strong, high grade timber; selected

sheltered forests in Wales can grow valuable, productive world class softwoods.

14. Understanding the role of different wood types and their properties, selecting them,

designing products accordingly and then marketing appropriately are the keys to realising

full potential of Welsh softwoods.

15. Simple protocols, ideal for SMEs, allow conversion and drying of high grade homegrown

softwoods.

16. The attractiveness, even beauty, of homegrown softwoods is under-appreciated. More

collaboration between designers and timber specialists could further differentiate and add

value to homegrown softwood products.

3

Introduction Welsh forests grow some of the largest conifers in Europe, see image 1 below.

Several well-known foresters and academics who have worked in Wales, for instance Charles Ackers,

Talis Kalnars and Roger Cooper have promoted Wales and western Britain as ideal for growing

conifers. Kalnars and Cooper have suggested that forest-based industries in Wales could play a

leading economic role with conifer forests as the new industrial resource. Furthermore the Welsh

environment could sustainably grow more high-yield conifer forests on currently underutilised

marginal land, for instance bracken infested land. Ackers quite clearly saw great potential for

homegrown Douglas fir comparing it to that grown in Washington and Oregon: There is no reason to

suppose that we cannot grow in our western counties equally good 80 year old crops (Ackers C. P.

1947, 109). However, it is still a commonly held belief that fast growth of conifer species in Britain

produces low grade timber (Bawcombe J. M. 2011, 2) and this has often been cited as the barrier for

using homegrown softwoods in construction and other value added applications. Sawn spruce from

the new generation of Welsh sawmills built during the 1980s to convert 20-30 year old thinnings was

notoriously unstable in drying. This was because of the high proportion of juvenile wood in the logs

(Edwards C. 2015) and juvenile corewood still governs the characteristics of British structural timber

(SIRT 2014); but the same issue arises with young plantation-grown timber all over the world. Fifty

year old sawlogs are commonplace nowadays in the UK and the sawmilling industry has matured;

BSW at Newbridge on Wye produces kiln dried, strength graded, planed softwoods for a demanding

construction industry. Informed foresters and wood scientists now consider sheltered parts of Wales

to be amongst the best places in Britain to grow conifers (Gardiner B. 2015).

Normative judgements in regard to wood properties are not useful. There is enough evidence within

existing wood science literature by the likes of Zobel and Buijtenen (Zobel B. J. and van Buijtenen J.P.

1989) or Haygreen and Bowyer (Haygreen J. G. and Bowyer J. L. 1996) to counter generalisations in

regard to wood properties regardless of where in the world forests are grown. We know that it is

possible to grow construction grade timber in Wales but at the time of writing the increasing

strength of the pound against the euro and the Swedish krona (“This Is Money” 2015) makes

imported construction grade softwood increasingly cheap compared with homegrown material. It is

technically feasible to use homegrown softwoods in various frame types, glulam and massive-wood

panel types (Dauksta D. 2013). However technical feasibility does not guarantee economic viability

when markets are driven by volatile currency exchanges. In order to counter the influence of

exchange rate volatility on uptake of homegrown softwood the Japanese government has adopted a

points system for timber housing whereby up to 300,000 yen (£1640 at 01/03/15) credit is made

available for house buyers who specify use of homegrown softwood. However no such process exists

in Britain as yet and any options for incentivising use of homegrown timber could be useful.

Different selection and processing of homegrown softwoods may offer significantly higher returns

than for existing markets. Species such as Douglas fir which are already valuable on global markets

may offer opportunities for differentiation and selling into markets with higher margins where

economic viability is not so dependent on the strength of the pound against other currencies. Wood

science defines and measures the material characteristics of wood types allowing informed selection

in order to optimise use of varying material properties appropriately. To add to information

contained in this report, readers can also access the USDA’s Wood Handbook freely available from

here; http://www.fpl.fs.fed.us/documnts/fplgtr/fpl_gtr190.pdf

4

Image 1: A massive Douglas fir in North Wales, around 1.2m DBH

5

A short history of debate about ‘quality’ Foresters of an earlier generation such as

Elwes, Craib, Chalk, Turnbull, Edlin, Ackers and Hiley wrote enthusiastically about the great potential

for plantation forests of exotic tree species. These foresters were well travelled and had a broad

overview of forestry. For them fast growth was not a challenge to production of useful, high grade

softwoods. Elwes travelled across Britain visiting every species of tree growing outdoors for his book

The Trees of Great Britain and Ireland, he also travelled abroad extensively (“Henry John Elwes”

2015). Chalk and Hiley were amongst the first to report on wood properties of fast grown Douglas fir

in Britain. Craib, Hiley and Turnbull studied and reported the properties of fast grown pine in South

Africa. However, whilst posting positive results from experiments in plantation forest management

British foresters were simultaneously accused of growing ‘rubbish’ because of the fast growth of

conifers in the UK (Hiley W. E. 1955, 159). The findings of both Hiley and Turnbull were debated from

1955 (Aldridge F. and Hudson R. H., 260–270) to 1958 with amongst others C. P. Ackers doggedly

supporting the notion that we can grow useful high value softwoods in Britain (Ackers C. P. 1956,

51–57).

Sixty years later the Romantic fashion for denigrating the ‘quality’ of wood from planted conifer

forests in Britain continues. The careers of several journalists, for instance Michael McCarthy

(McCarthy M. 2010) and more recently George Monbiot have been bolstered by their anti-conifer

forest stances. The consequence of this drip feed of negativity is that even within the modern

forestry and sawmilling industry the belief persists that we cannot grow high grade conifers in the

UK. Contemporary foresters are forced reactively into justifying the existence of planted conifer

forests in Britain. However, recent political moves to implementing a ‘low carbon’ economy by using

sustainable resources has revived government interest in and the reputation of high-yield conifer

plantations somewhat, especially in regard to their capacity to sequestrate CO2 which can then be

stored within the fabric of timber structures and products.

The term ‘wood quality’ is often used and but rarely defined by its users. This term is generalist and

generally unhelpful when describing particular types or variations of wood in order to ascribe

appropriate applications to them. The term ‘wood qualities’ is better because it takes into account

the wide variations in properties found within tree stems, between trees, between forest stands and

between zones in which they grow. Wood qualities need to be defined in terms of end use and

fitness for purpose otherwise wood quality becomes an arbitrary term (Larsen P. R. 1969, 1,2).

Indeed because quality is often used as a judgemental weasel word, use of the neutral term ‘wood

properties’ might be more appropriate. Users need to understand that timber from plantations is

different from that of old growth forests but not necessarily of ‘bad quality’. Fitness for end purpose

is the only absolute criterion for selection of timber.

A 1960 article in Unasylva recounts the historical experience of New Zealand foresters with Pinus

radiata. It is pertinent to the 60 year old ‘quality’ debate continuing here in Britain now:

The scale and speed of exotic planting have been so great in New Zealand that there have been

acknowledged mistakes, but it is interesting to find how much success has been achieved.

The exotic planting boom of 1923-36 led to serious mistakes and financial losses. It is wrong to

attribute those losses to faults in the chief species concerned, P. radiata. The losses were due to

mishandling. Subsequent successes have proved what an excellent tree is this pine when well

6

handled, and how productive it is of good lumber and pulp if so handled. The New Zealand Forest

Service considers among the major mistakes that were made in planting exotics were unsuitable

sites, inferior seed and planting techniques, lack of tending (weeding, thinning), too wide spacing,

inadequate fire protection, a bad age-class distribution, and, in some cases, poor location in relation

to markets. This is a formidable list on which it is easy to be wise after the results are apparent.

Furthermore with a degree of prescience in regard to current debate about design of future

plantations the article says of mixed-age forests:

Those who advocate uneven-aged forests of intimately mixed species must take account of the

strong preference of manufacturers for considerable coupes of one species and age to simplify and

concentrate extraction work and give as much uniformity as possible to the batches of material

arriving at any one time at the pulp mill or sawmill. It may be necessary to compromise between the

planting of different species according to the varying sites available and the economic advantages of

uniformity to the manufacturer so as to secure the optimum, avoiding both unhealthy or inadequate

crops and undue processing costs (Scott C. W. 1960).

Problems associated with establishment and perception of conifer plantations in Britain were

discussed by W. L. Taylor in Journal of Forestry in 1958. The author blamed dogma and inelastic

formulae for harming conifer forestry in Britain; it was authoritatively spread around that we could

not grow quality conifers in Britain. However, it was clear at the time that industrial plantation

forestry was so new to this country that foresters could not possibly know enough about the

complex interactions of seed quality with provenance, local conditions and the many other factors

which influence forest growth (Taylor W. L. 1958, 22–24). Hindsight is easier to use than foresight

and the former is utilised by those detractors without the foresight and imagination to see the

potential for conifer plantation forests. Taylor sums it up succinctly: Softwoods are among the most

versatile of natural products. Surely we grow our conifers to enable us to make the fullest use of this

versatility (Taylor W. L. 1958, 22).

Wales has highly varied terrain, microclimate and soil types. Anything that can change tree growth

patterns may cause wood variations (Zobel B. J. 1992, 31) and so the variability of the Welsh

landscape gives huge potential to grow a wide range of softwood types even within a single species,

see image 2 below. It is possible to find Welsh softwoods with growth rings as small as 1mm wide or

as large as 25mm wide, the larch on the right of image 2 has growth rings up to 15mm wide, the

slow grown larch on the left has some growth rings around 0.5mm wide. However, commodity

timber processors and even many smaller processors working with Welsh grown softwoods

currently do not select in order to further valorise Welsh softwoods and do not understand the

properties of varying wood types in order to match them with appropriate products. This is hardly

surprising; very little study of wood variation is carried out in the UK. Wales’ situation is greatly

different to that in Scotland where there is a much larger conifer forest resource which may produce

10 million m3 or more by 2020 (Moore J. et al 2009, 384) and Scottish processors will be trying to

position high volumes of spruce competing directly with commodity softwoods on global timber

markets. England is different again in being dominated by unmanaged broadleaved forests which

make up 70% of forest area. Wales is in a unique position within the UK; owning a small, manageable

plantation forest area with great potential for partial transformation to growing high value

softwoods selected for appropriate high value applications. Trials with specialist processing and

7

marketing of existing minor softwood species and oversize logs of principal species could allow SMEs

to develop bespoke, branded products which are sold through differentiation rather than by lowest

price in already crowded markets. We need to ‘get over’ old negative perceptions of conifer

plantations, look again at the wood we grow and market it with a positive attitude.

Image 2: 47 year old Japanese larch (left) & 15 year old (right) grown on Welsh sites 3km apart

Welsh forests cover around 300,000 ha, the management of which suffer from conflicting societal

demands, expectations and aspirations. They comprise roughly half conifer, half broadleaved

species. Only around half of the forest is managed and productive; this is the plantation conifer

forest which is composed largely of exotic species brought originally from the Pacific north-west

coast of America by the Scottish plant collector David Douglas in the early 19th century. We still have

a limited understanding of the wood properties of homegrown softwoods. No large scale systematic

study of wood properties between sites had been undertaken in Britain before the 2007 Sitka spruce

benchmarking project led by New Zealand forester John Moore (Moore J. 2008, 12); the final report

was only published in 2011. Information on other UK-grown conifer species is so scanty that

researchers still refer to Gwendoline Lavers The strength properties of timbers published nearly half

a century ago. Anyway because of changes in forest management practices some commentators

suggest that historical data on wood properties may not necessarily be relevant (Moore J. et al 2009,

384). 57 years ago W.L. Taylor considered the renaissance of forestry in Britain to be in its early

stages (Taylor W. L. 1958, 22). Although progress has been made especially through the efforts of

researchers working in Scotland, we still lack basic data for many species. Zobel makes it clear that

the only way to know what type of wood will be produced by exotic species in a particular

8

environment is to grow them there and then study their timber properties (Zobel B. J. and van

Buijtenen J.P. 1989, 56). However, the type of wood science that we need to catalyse ‘a renaissance

of British forestry’ is under-represented, and severely under-funded.

Douglas fir (arguably one of the most valuable conifer species growing in Britain) has been studied

even less than Sitka spruce. The most comprehensive recent work was done by Jonathan Bawcombe

and published as a thesis in August 2012; however this project only covered Douglas fir grown in

southwest England. The UK-wide study of Douglas fir undertaken by Tom Drewett under the aegis of

SIRT (Strategic Integrated Research in Timber) based at Edinburgh Napier University is due for

publication by mid-2015. Despite the limited amount of recent research carried out on the

properties of homegrown conifers, it is becoming clear that parts of western Britain are capable of

growing softwoods comparable to plantation timber grown world-wide; Douglas fir being a case in

point (Bawcombe J. M. 2011, 18). Jonathan Bawcombe’s thesis is available here;

http://opus.bath.ac.uk/32245/1/UnivBath_PhD_2012_J_Bawcombe.pdf

High mean wind speeds across Britain (Met Office 2014) seem to strongly influence softwood

properties not only by formation of spiral grain (Fonwenban J. et al 2013, 1 ) and reaction wood but

also possibly affecting stiffness of UK softwoods in other ways (Mclean P. 2014). However many of

the planted species, but most especially Sitka spruce and Douglas fir, originate from locations with

high average wind speeds anyway; that is the Queen Charlotte Islands and the Rocky Mountains

(“GLOBAL WIND MAP” 2015). Arguably both species are already well adapted to growing in windy

zones so there is a question as to why British conditions might adversely affect those species any

more than conditions in northwest America. However thigmomorphogenetic theory as proposed by

F. W. Telewski and others states that plants respond to wind and other mechanical perturbations in

a way that allows continued plant survival in windy environments. Studies have shown that induced

flexure causes both increased stem elasticity (lower MOE) and lower flexibility (Telewski F. W. and

Jaffe M. J. 1986). Through adaptive growth, stems can also form elliptical cross sections along the

same axis of prevailing wind and stem height may be reduced (Mattheck C. 1998, 26). The siting of

forest plantations on lower value marginal land often in exposed uplands in Britain where declining

agriculture offered opportunities for afforestation has influenced mechanical properties of the trees

either through thigmomorphogenesis, reaction wood formation, spiral grain formation or

combinations of these processes. Despite these challenges Britain has developed a significant,

modern softwood processing industry capable of supplying commodity timber markets. Sheltered

forests growing higher grade softwoods could supply higher value bespoke markets.

Wood is a remarkable material which varies greatly; compare for instance the wood properties of

very low density balsa compared to those of very high density lignum vitae. Over millennia people

have utilised the widely varying properties of wood to make an enormous range of products and

artefacts from the smallest and most humble of objects such as wooden spoons to some of the

largest structures and machines on the planet e.g. the Tacoma dome in America, the Todaiji Temple

in Japan (see image 3 below) and the ‘Spruce Goose’ aeroplane built by the Hughes Aircraft

Company at the end of WW2. At the other end of the scale, high-specification nanocellulose

products are being commercialised right now. As the structure of Welsh conifer forests change in

order to accommodate societal expectations, climate change and policy change, there will be a

wider range of wood types available for future generations; especially if more species and/or age

diversity is implemented. In order to make best use of wood grown in Wales and foster innovation,

9

we need to gain a better understanding of its varying properties. Armed with more knowledge,

better strategies can be formulated and informed product development can be carried out leaving

no excuse for attempts at ad hoc research and design projects which rarely lead to successful

commercialisation. In order to describe and use wood adequately and efficiently we need to know

and quantify several of its properties, therefore the word ‘quality’ is of limited utility.

Image 3: Todai-ji in Nara, Japan; one of the largest timber structures in the world

The report Welsh Softwoods in Construction which was published in December 2013 analysed and

set out how Welsh-grown softwoods could be used in construction. The intention of this current

report is to use existing wood science literature to develop the notion that not only can we build

with Welsh softwoods from conifer plantations, but we can also use them in higher value added

joinery and other applications. This study will question commonly held beliefs about timber

characteristics and re-examine three plantation conifer species grown in Wales. A rational, evidence-

based process will be followed to understand what properties are useful to create successful added

value products such as joinery and appearance-grade softwood products from Welsh conifer

plantation timbers. The results of some limited practical tests will also be available in appendices.

Total UK consumption of timber and panel products by value was £3.1 billion in 2013. At 8.5 million

cubic metres per annum softwood is by volume the single largest wood product consumed; its value

around £1.45 billion (TTF 2014, 24). About 10 % of all sawn softwood consumption in the UK is used

in joinery such as windows, doors and stairs (NBS 2014). Therefore the value of the UK softwood

joinery market is around £145 million. There is already a large market for softwood products in the

UK using a wide range of products and homegrown softwoods make up 42.5% of UK consumption by

volume (TTF 2014, 6). Is there potential to select some higher grade softwoods in order to position

10

them in higher value market segments? Although the joinery market makes up only 10% of

sawnwood consumption, there may be more potential for local SMEs to realise value in comparison

with the small margins gained in modern high volume timber frame manufacture (Mclean P. 2014).

Two of Wales’ most modern timber frame manufacturers have been forced into administration over

the past year, the most recent being ‘Framewise’ in Presteigne (Prior G. 2015).

Variability in trees The importance of variability in trees cannot be overstated and

generalised statements about wood properties of any particular species are at best unhelpful. Tree

to tree differences within a species, within a stand or within a provenance are large. The variation of

wood properties within one species can be greater than between species and it is necessary to

sample at least thirty trees to obtain a valid estimate of wood properties within any given stand. For

instance Radiata pine trees grown on the same site can exhibit up to 60% difference in wood density

and there can be more differences within one site than the average differences between sites.

Within any one tree variations in properties can be greater than differences between the same

species grown in the same environment (Zobel B. J. and van Buijtenen J.P. 1989, 73–77). Zobel sums

up:

1. Tree to tree variation is so large that it is of major importance for all wood products.

2. The magnitude of variation within individual trees makes it difficult to distinguish site,

environmental or silvicultural effects on wood properties.

3. The amount of tree to tree variation differs considerably between species; some species can

have more consistent wood properties than others.

4. Earlywood and latewood properties differ greatly and their relationship alters factors such

as machinability, latewood properties greatly influence final product quality.

5. Much tree to tree variation is genetically controlled.

6. Use of vegetative propagation can produce trees with similar wood within clones but wide

variations between clones.

7. High variability between trees and strong genetic control makes for very successful

improvement of wood properties.

8. Foresters, processors and researchers need to recognise the variability of wood, learn to live

with it and learn how manipulate it.

Variability caused by geography and provenance is summed up thus (Zobel B. J. and van Buijtenen

J.P. 1989, 46–47):

1. Wood grown at higher elevations has lower density (and stiffness) than wood grown at

lower elevations.

2. Wood grown at higher latitudes has lower density (and stiffness) than wood grown at lower

latitudes; there is a strong tendency for increased density (and stiffness) from northeast to

southwest.

3. Wood grown at higher latitudes generally has smaller knots because of smaller limb size and

less degrade due to crooked trees.

4. Wood density can be higher in maritime conditions compared with drier inland zones.

5. In warmer climates with extended growing seasons where rainfall is high in late summer and

autumn, conifers produce more latewood and therefore have a higher wood density.

6. Wood of the same species within its natural range can vary significantly.

11

Points 4 and 5 suggest that the Welsh climate can benefit some wood properties of conifers. Point 1

suggests that slower grown, lower density timber may be grown in sheltered areas at higher

elevations. However effects of latitude and elevation are complex and should not be considered

separately (Dykstra D. P. and Monserud M. A. 2007, 116–117) but anyway sourcing and selection of

material from higher sites in Wales is relatively easy and might prove rewarding.

Much research has been carried out across the world on methods for overcoming defects in timber

which could have been minimised by different forest management. There is huge scope for genetic

improvement of timber through tree-breeding because most wood properties are so strongly

inherited. Tree form, growth rate, wood morphology and wood chemistry can all be manipulated to

improve wood but efforts so far have tended to be focussed on fast growing, pest resistant trees

despite the desirability of high grade timbers (Zobel B. J. and van Buijtenen J.P. 1989, 249).

Management of wood properties will increase forest product quality and the concomitant potential

for manufacture of quality products (Zobel B. J. and van Buijtenen J.P. 1989, 303). Better knowledge

of our softwoods and matching wood types to quality products gives scope for more effective

marketing, focussing on positive attributes rather than lowest price. Different wood types arising

from tree to tree variations and within-tree variations can be selected during processing in order to

better utilise particular wood properties. This approach is different to that of machine strength

grading where a strength class will be assigned from a spread of values analysed by statistical

procedure. Wood variation allows choice of particular properties and selling points from a range of

grades or classes.

Better knowledge of how wood properties might be optimised for particular applications can inform

the design and future management of high yield Welsh forests. The huge range of variations in

wood properties clearly points to the need for selection of different grades or types of wood in order

to optimise their utilisation. Quality is often used generically to describe timber but is generally

meaningless in this context because users and specifiers wish to understand whether a particular

wood type is fit for their desired application. In Wales, quality is often used as a weasel word in

order to reinforce negative views about homegrown softwoods. Quality is better used to describe

wood products, not wood types. We need a wide range of descriptive terms to match the wide

range of wood types encountered in timber sourced from Welsh plantation forests.

Planted forests Not only do we need to understand wood variability better here in Wales; we

also need to better understand the role of the conifer plantations where industrial timber is grown.

Planted forests are now considered to be foundations for creating sustainable economies. A recent

report which is available from the Food and Agriculture Organisation of the United Nations (FAO)

attempts to quantify the growing importance of planted forests around the world and clearly

demonstrates that global industrial roundwood needs can be met by using a small proportion of the

world’s forested area for growing high yield plantation forests. The figures are compelling; global

area of plantation forests increased from 178 million ha in 1990 to 264 million ha in 2010, making up

7% of total forest area. Although only covering this small proportion of the planet’s forest area,

estimates show that plantation forests supply between one third to two thirds of global industrial

roundwood demand and sequester 1.5 gigatons (1.5 x 109 tons) of carbon per year (Mendes A. et al

2013). In other words, using the worst case scenario, current global industrial roundwood needs

could be met from plantation forests grown on only 20% of the world’s forest area. Using the best

case scenario it may be possible to meet global needs on only 10% of the world’s forest area.

12

The report Planted forests are a vital resource for future green economies summarises the results

from the 3rd International Congress on Planted Forests and is available here;

http://www.fao.org/forestry/37902-083cc16479b4b28d8d4873338b79bef41.pdf

This report states: The contribution of planted forests to addressing the major socio-economic and

environmental challenges of our time - poverty alleviation, food security, renewable energy,

climate change and biodiversity conservation- is widely acknowledged. In many developing and

developed countries planted forests have become a substantial component of the productive and

protective forest resources. The congress noted in this context that statistics on wood production

from planted forests including small woodlots and trees outside forests are incomplete and tend to

overlook local markets for fuelwood, poles and other minor wood products. Globally, extensive areas

are estimated to be available for forest restoration, rehabilitation or reforestation. However

the report also notes that in Europe there is a growing mismatch between actual societal demands

and forest policy: Current trends in European forest management could result in an over-supply of

wood from broadleaved species, as well as a shortfall of coniferous timber from European forests, in

particular as an increase in harvest is difficult to achieve due to restrictive environmental policies. On

the other hand in Asia planted forests are perceived as a force for good: The National Development

Strategy of China (2013) emphasizes the establishment of an Ecological Civilization. Planted forests

are increasingly assigned to protective and multiple-use functions with the objective to optimize land

use at a landscape level.

Researchers are aware that high yield plantation forests need to be redesigned in order to gain

wider acceptance: Large-scale single species plantations with strong environmental impacts should

be replaced by more ecological and integrated approaches at stand and landscape levels, including

mixed species plantations, and greater use of silvicultural systems based on close-to-nature

principles. However: Planted forests are exposed to socio-economic risks due to governance failures.

These risks comprise a weak or inadequate forest policy framework including insecure investment

conditions (Mendes A. et al 2013). Forests with varied management and structures can anyway

contribute significantly to biodiversity whilst providing overall fossil fuel savings with up to 37%

reduction of CO2 emmissions (Dearing Oliver et al 2014, 268–269). If management of plantation

forests in Wales moves towards close-to-nature methods and if industrial forests are to be

diversified by age and/or species mixes then even greater wood property variations can be

expected. Therefore Welsh Government’s own forest diversification policy implies a future need for

increasing the study of wood science within Wales in order to better utilise the wide range of

softwood types. Research will need to keep pace with policy implementation.

Computer modelling suggests that there is no conflict between the goal of stand diversity and the

goal of wood quality. Stands with a higher diversity of structure (tree size) and a higher diversity of

species may even contain a higher percentage of high-grade logs (Liang J. et al 2009, 173). However,

UK researchers have suggested that forests under conversion to continuous cover type forests with

significant numbers of edge trees around felling coupes may increase formation of undesirable

reaction wood in those edge trees because of heavy branching on the light side of coupe margins.

Even less desirable light-seeking leaning trees may also grow in these margins thus increasing the

proportion of low grade logs within continuous cover stands (Macdonald M. and Gardiner B. 2005).

This concern may only apply to stands during the conversion phase to continuous cover

management, because by definition continuous cover silviculture when mature implies vertical

13

closure and no edges within forest stands. Until more continuous cover silviculture is carried out and

studied in Wales, this topic can only be debated. Some of the factors are set out below in this table

taken from the paper Trees for the Future (Macdonald E. et al. 2008, 8).

Advantages

Disadvantages

Planting

1. Opportunity to select species

2. Volume and quality gains from

improved progeny

3. Control over stocking

4. Minimal variation in age class

structure resulting in more uniform

growth and timber properties

1. Possible stem form problems

associated with nursery practice

and early instability (toppling)

(Watson and Tombleson, 2002)

2. High establishment costs

3. Possible patchy stocking – future

timber quality problems

Natural

Regeneration

1. Improved stability – possibly better

stem form

2. Potentially high stocking and large

number of trees for selection in

thinning

3. Potentially low cost

4. Fits well with Continuous Cover

Forestry

1. No opportunity for improvement in

growth or timber quality through

use of selected

provenance/progeny

2. Difficult to control species mix –

may have a lot of low value

species (e.g. hemlock, grand fir)

3. Costs of respacing/pre-commercial

thinning

Table 1: Advantages/disadvantages of planted/naturally regenerated forests (Macdonald E. et al.

2008)

As table 1 suggests, well-managed even aged plantations (which may also benefit from an

understorey) allow production of more consistent or uniform sawlogs demanded by modern

processors. High-pruning and other silvicultural interventions can help trees to form higher grades of

timber. However this approach lacks the ‘green’ storyline of continuous cover forestry which in

Germany has the more romantic name of dauerwald or eternal forest. Anyway, selected British

conifer plantation forests managed for higher value timber could surely benefit from an understorey

regardless of the silvicultural system employed. It may be useful to study the extensive literature

available from the United States Department of Agriculture (USDA) on the topic of Pacific north-west

coast old growth forests. The mixture of natural and anthropogenic processes and events which

have shaped those forests may give clues as to how we get better results from our own forests.

Underplanting Douglas fir with species such as Western hemlock or Western red cedar imitates

nature (Franklin J. F. and Spies T. A. 1983) whilst achieving results such as reduction of branchiness

which otherwise requires high pruning. Underplanting with beech or ‘native’ alder might also

achieve the same objective whilst improving ‘green’ credentials of plantation forest.

14

One aspect of growing multi storeyed forests is clear; trees with desirable or marketable properties

selected for long term retention will benefit from reduction of side branching and creation of a

beneficial micro-climate within a shade bearing understorey. Western hemlock is currently an

unfashionable species blamed particularly for its invasiveness. This perception could be turned on its

head; natural regeneration is easily achieved with Western hemlock and as a shade-bearer it is ideal

for continuous cover and multi storey forests. In order to justify any extra management input,

timber from more intensively managed stands will need to be sold into higher value markets.

Therefore we need to know more about the material characteristics of species such as Western

hemlock in order to optimise marketing. Sixty years ago Ackers wrote: The timber from this tree is

now so very well known in our import trade that I hardly need enlarge on its importance.

Homegrown timber is getting a good name (Ackers C. P. 1956, 53).

British users have forgotten how desirable joinery grade Western hemlock timber has been to UK

users, until recently making up to £700/m3 (Capricorn Eco Timber 2013) although actually it was a

relatively new industrial species for both British and American markets. In 1908 it made up only 0.2%

of output in America and by 1926 it was the third most important species. However, because it is

shade-bearing and consequently not self-pruning, its persistent branches will lower ultimate yields

of clear timber compared with Douglas fir and Sitka spruce when converted. On the other hand,

common defects which frequently degrade other species occur less with Western Hemlock; for

instance, dead knots, spiral grain, resin ducts or pockets and blue stain are all less common

compared with its associated species (Johnson R. P. A. 1929, 5–8). However both in Britain and its

native range stem fluting is a well-known problem. Fluting is associated with fast growth in Western

hemlock. Narrow spacing, slower (sheltered) growth conditions and shorter rotations reduces fluting

(Singleton R. et al. 2003, 221–228). Therefore growing it as a ‘catch crop’ under large valuable trees

such as Douglas fir could optimise growing conditions for both species.

According to G. M. Lavers, homegrown Western hemlock is of the same stiffness as Sitka spruce

(Lavers G. M. 1969, 16–20). American literature suggests that Western hemlock has some excellent

working properties; it has very high shock resistance for its weight and hardness is moderate.

Perhaps most importantly its hardness within growth rings is relatively uniform (falling somewhere

between hard and soft pines); giving it even-wearing and good shaping and machining properties

(Johnson R. P. A. 1929, 22–26).

Branchiness in Welsh plantation forests is perceived badly by both foresters and the public (Woods

R. 2015); foresters see it as indicating low grade timber and the public have difficulty in seeing any

amenity value in the dark, branchy spaces between rows of trees. However USDA literature clearly

demonstrates that side branches also grow thickly in naturally regenerated even aged Douglas fir

forests in Washington and Oregon, see image 4 below. This is a passing phase and self-pruning

occurs as stands age (McArdle R. E. and Meyer W. H. 1949, 4–10). Size, location and frequency of

branches are seen as major wood quality factors. However because plantations here are generally

not retained after year 50 and we know that self-pruning does not occur until around year 67 in

Pacific NW America (Gartner B., Robbins J., and Newton M. 2005) we do not normally see this

transition in our plantations. Therefore in Britain our plantation trees are often perceived as too

knotty to make higher grade timber whilst actually they are too young to have formed thick layers of

clear wood over the knotty core; as well as species provenance and tree breeding, tree spacing and

rotation length are key factors in improving out-turn of high grade wood. Longer rotations have

15

already been suggested as a possible solution in order to increase yields of strength graded

construction timber from mature heartwood of Sitka spruce and reduce rates of rejection because of

distortion. However, windthrow will be a limiting factor for many sites (Macdonald E. et al. 2008, 6–

9) therefore long term retention can only be a solution on selected, more sheltered sites.

Researchers working in Scotland under the Strategic integrated Research in Timber project or SIRT

have already pinpointed the factors influencing wood properties from conifer plantations in Britain.

Pragmatic decisions for gaining higher grade timber from industrial forests in Wales could already be

made on the basis of work done in Scotland as well as that done by Zobel.

In Trees for the Future the writers conclude: Silvicultural intervention can play a key role in

determining the quality of timber produced from conifer forests in the future. The outcome will not

depend on any one action – impacts on timber quality depend on every aspect of management. A key

priority must be commitment to the production of quality timber at each stage: “from plant to

plank”!(Macdonald E. et al. 2008)

Image 4: An old image of heavily branched trees in a 50 year old Douglas fir stand in PNW America

Perhaps selected sheltered plantations could be subject to long term retention in order to study

stem improvement with age. Long term retention allows thicker layers of high grade clear material

to grow around the juvenile core and knotty inner heartwood. This then implies the need to

separate high grade clear timber from lower grade knotty or juvenile timbers by specialist

processors. Long term retention of Douglas fir along with understorey planting can create a forest

16

with old growth forest features and a choice of harvesting options but increased management cost

will need concomitant increased value. C. P. Ackers calculated that clear timber made up 15% or less

of the total crop from old growth forests in PNW America and that in Britain with correct

management or possibly with high pruning we could do better (Ackers C. P. 1947, 111).

Conifer plantations supply almost all of Wales’ industrial roundwood needs from only 7% of Welsh

land area and half of Wales’ total forest area; the other half of forest area is made up of largely

unmanaged native broadleaved forests. Forest industry representatives are concerned that this

small productive conifer forest area may suffer production shortfalls in the future because of weak

or inadequate implementation of a pragmatic forest policy. Furthermore there is a perception within

the forest industry that some Welsh government policymakers still aspire to create more non-

productive native broadleaved amenity forest at the cost of softwood production. According to

Martin Bishop of CONFOR; planting more broadleaved forests will create yet more unmanaged forest

in Wales (Bishop M. 2014). However the WWF is predicting that under their ‘Living Forests Model’

global demand for industrial wood could triple by 2050 (WWF 2013, 8). If this is the case, under the

present planting scenarios Welsh conifer forests may not even be able to supply one third of Wales’

own industrial roundwood demands in 2050 at a time of unprecedented global demand for timber.

In this case there will be demand for all grades of wood from juvenile to best grade clears.

The WWF report suggests that increased demand should be met by both plantations and increased

management of natural forests (WWF 2013, 2). However in Wales almost all current forest

management and timber production is carried out in plantation forests. This report is specifically

intended to counter claims that Welsh conifer forests mainly produce ‘low quality’ or ‘low density’

sawlogs. Actually wood science demonstrates that we can grow exotic conifer species with

properties equal to the same species growing in their native ranges. Indeed, Japanese larch timber

grown in Wales appears to demonstrate somewhat higher stiffness (MOE) than that grown in Japan

and the recent bending tests carried out at Napier University show it is also on average slightly stiffer

than Scottish-grown Japanese larch. However, although the results from Napier University were

statistically significant, the methodology does not allow firm conclusions to be made (Ridley-Ellis D.

2015). Nevertheless it is a recorded phenomenon that softwoods tend to decrease in density and

stiffness with increasing latitude and altitude (Zobel B. J. and van Buijtenen J.P. 1989, 196). Wales

has a rainfall distribution which favours some conifers and its mild autumns allow extended growth

periods at the time of year that conifers lay down the denser, stronger latewood zones of growth

rings. In other words the Welsh environment has distinct, positive attributes for growing high grade,

high stiffness softwoods.

Our main problems are perception and misinformation. Actually we can make better use of the

timber grown in our highly productive plantation forests because we already grow conifer trees with

a range of appropriate properties for use in high value added products. However the Welsh forest

industry mind-set has largely been stuck in a commodity culture based around processing Sitka

spruce for high volume markets and is incapable of realising the full value of sawlogs which fall

outside the volume processors’ specifications. For instance ‘oversize’ logs which are too big for

modern mills such as BSW Newbridge have been considered to be a problem rather than a different

type of (more valuable) asset. Smaller sawmillers seeking large diameter Sitka spruce in the past

have done so not because of the potential for processing differentiated products but simply because

they are unwanted by the volume processors and have consequently been cheaper. As availability of

17

Sitka spruce has declined in recent years, larger sawmillers have been forced to take in higher

volumes of other species, e.g. larch from Phytophthora ramorum affected forests. However, larger

mills have preferred to substitute this larch for spruce within their existing markets rather than

develop new markets. Despite its potential for niche construction markets and availability of new

machine grader settings, most high stiffness construction grade larch is likely at best to be sold as

C16 material by volume sawmills. To sell the strongest softwood grown in Britain in this manner

appears to demonstrate a somewhat risk-averse marketing culture. The Japanese (both historically

and recently) built some of the largest timber structures in the world with their own timber. So far

the British are (with a few honourable exceptions) still making fences with their own timber. At best

the British are just starting to use homegrown C16 strength graded softwoods in low rise timber

framed construction whilst somewhat contrarily they built the world’s first modern nine storey

timber structure using imported timber panels (Waugh Thistleton Architects 2015).

In order to supply increased future demands for industrial roundwood, the proportion of wood

grown in high yield plantations will continue to increase globally and societies will need to use it

more efficiently. We can use wood science to counter the misinformation surrounding effects of

growth rate on wood qualities. There can be high and low density woods in equally fast growing

trees and within individual trees; however the wide range of published literature has been cherry-

picked to support many points of view (Zobel B. J. and van Buijtenen J.P. 1989, 157). Anyway we will

need to use land more efficiently because they’re not making it any more (Goldin I. 2014, 184).

Therefore high yield plantation forests are the rational choice for sustainably growing more timber

on a crowded planet. Furthermore growing conditions and silvicultural interventions can be better

managed within plantations in order to optimise response of trees and influence timber qualities.

Exotic woods grown in forests away from their native habitats may exhibit different characteristics

to those grown in their original environments but these differences do not necessary make the

woods inferior (Zobel B. J. and van Buijtenen J.P. 1989, 58).

A modern, large scale forest product industry has been developed in New Zealand through

continuous investment in wood science and silvicultural research. In a country with only slightly

larger land area than the UK, forest cover in New Zealand now stands at around 31% of land area

with conifer plantations covering only 7% of total land area. The planted forest covers just over 1.7

million ha (with 1.5 million ha of Radiata pine) producing 23 million m3 of roundwood. Forest

products now contribute 3% of New Zealand’s GDP; they are the country’s third largest merchandise

export and the industry has a target to double export earnings by 2022 (NZFOA 2013). Amongst the

researchers now working there is Bangor University alumnus Addis Tsehaye who studied variations

in wood properties of Radiata pine grown in New Zealand, contributing to the body of data which

underpins the industry’s progress (Tsehaye, A. 2015). Fast grown, wide ringed Radiata pine seems

like an unlikely candidate for manufacture of quality products but research demonstrated its

potential and New Zealand joinery products are now marketed widely in the UK.

18

Log variability Wales produces some of the largest Douglas fir grown in Europe and despite

the fact that this is a valuable niche market species, in the past most large Douglas fir sawlogs have

been sold to specialist sawmills in England thus losing potential local economic gain for Wales. There

has been a failure to recognise the value of specialist softwood sawlogs in Wales. Few SMEs optimise

income from them through better matching to bespoke products and subsequent product

marketing. This failure has reinforced negative perceptions about growing and selling large diameter

softwoods in Wales. A self-fulfilling negative culture is shared by foresters and sawmillers alike.

Indeed, sawmillers are often happy to denigrate specialist timber grown in Wales simply in order to

drive down log prices, leading to more hand-wringing by foresters only seeing opportunity cost and

consequent folly in growing big conifers and then selling them too cheaply. Assumptions and

accusations such as those made by Aldridge and Hudson in the 1950s about fast growth of species

such as Douglas fir persist and prices are adjusted downwards accordingly. However, specialist

timber merchants ‘Capricorn Eco Timber’ near Stafford no longer discriminate between homegrown

and north American Douglas fir; they sell merchantable grade material from both sources at the

same price around £600/m3 in 2013. Furthermore Capricorn make the point that old growth Douglas

fir is no longer available anyway so even imported clear material is from second growth or planted

forests. With imported high grade Douglas fir being sold at £1000/ m3 (same price as joinery oak) in

2013, it is now worthwhile to select higher grade homegrown material for sale into the same value

added markets (Capricorn Eco Timber 2013). High grade Douglas fir is grown in Wales but rarely

processed here in any volume. Very high grade Douglas fir from North Powys is shown in image 5

below. Juvenile corewood extends over nine or ten rings and is hardly discernible in this image. The

growth rings in the mature heartwood vary between around 2.5mm to 3.5mm in width, in line with

European timber trade descriptions of ‘high quality’ (“Russian and Latvian Sawn Timber” 2015)

Image 5: Very high grade Douglas fir from North Powys

19

Image 6 below shows Welsh-grown Douglas fir timbers from two different sites with obvious ring-

width differences and different wood properties. The wide latewood rings in the large dimensioned

stock from West Wales suggest high rather than low density despite the obvious fast growth of the

tree. Both timbers could be described as high ‘grade’ or ‘quality’ but those descriptions do not

convey the information needed in order to make informed decisions about suitability for particular

applications. Large diameter Douglas fir sawlogs can yield very wide stable boards which are not

available in other common softwood species. There is an untapped potential for selling material of

this type into high value specialist niche markets.

Image 6: Douglas fir from two different Welsh sites with large variations in wood properties

Suppressed, ‘drawn’ small diameter softwoods grown in Wales have also been seen as problematical

and are often undersold although their wood properties are not dissimilar to the small diameter

slowly grown softwoods of the Baltic region. Nordic and Baltic producers make a point of marketing

the wood properties of their slow grown softwoods, for instance: Rather severe climatic conditions in

Russia causes quite slow wood growth that leads to the narrow-ringed structure of timber. According

to the European standards the annual ring growth for the high quality coniferous timber should not

exceed 4mm. Russian timber in general has annual ring growth about 1.1-3mm (“Russian and Latvian

Sawn Timber” 2015). Welsh processors have so far not recognised the potential to properly valorise

small diameter softwoods. Indeed because small diameter juvenile thinnings have been in short

supply for many years in Wales, suppressed small diameter conifer timber with narrow growth rings

(a completely different product to juvenile thinnings material) has been routinely processed into

fencing products. Image 7 below shows a board cut from a suppressed homegrown larch tree from

a 47 year old plantation, after edging its dimensions were only 50mm wide and the smaller growth

rings are only just over 1mm in width.

20

Image 7: a board cut from a suppressed homegrown larch tree with narrow ring width

If there is a degree of hand-wringing at our lack of creativity and poor marketing skills, there is also a

degree of finger-pointing by those commentators who would prefer to see the end of softwood

production in Wales. If Welsh Government is to make pragmatic choices for the future of Welsh

forests then there is a need to break out of the negative circular thinking which dominates the

industry and the micro-politics surrounding it. Sawmillers routinely denigrate homegrown softwoods

in order to reduce log prices; Welsh softwoods achieve some of the lowest roundwood prices in

Europe not because of ‘quality’ issues but because of the vicious circle of misinformation which

surrounds the timber processing industry in Britain. The wide range of wood types grown in Wales,

especially those produced by very fast and very slow growing conditions are not being sold into

markets which reflect their special properties.

The Welsh forest has been shaped by the Industrial and Agrarian Revolutions and two world wars. Its

future size and design has not yet been decided and current debate is informed not only by rational

analysis of future needs of the sustainable society to which Welsh Government aspires but also

romantic ideologies which still tend to focus on planting native broadleaved forests. There is a wide

spectrum of services expected of forests grown in developed western societies and an aspiration

that all forests be multi-purpose. However there are few mechanisms developed which create

incomes to offset the extra costs incurred. It is therefore logical for now to focus on the

management of commercial exotic softwoods grown in Wales in order to optimise incomes from

21

them. Selecting, grading and marketing them more intelligently will generate more income through

positioning selected softwoods in value-adding niche markets. Welsh processors are already good at

converting homegrown softwoods for selling in commodity markets. A modern, productive conifer

processing industry has been established in Wales which is capable of partly servicing the demands

of a sustainable construction industry. However we have only just started to realise the potential for

marketing specialist or bespoke softwood products. If marketing is the vital tool for placing timber

products in the modern marketplace, wood science is the principal tool for understanding those

materials which we wish to market better.

Wales has an ideal environment for growing conifers but there has been a misunderstanding about

the role of planted conifer forests all across Europe. Ignorance, opinion and polemic have created a

self-perpetuating negative mythology which has stalled expansion of conifer plantations. Gerald

Ratner notoriously sabotaged his family’s jewellery business by denigrating the products sold in his

shops. Uninformed commentators have inaccurately and unfairly used the ‘Ratner word’ to describe

Welsh-grown softwoods reinforcing the myth that Welsh conifer forests only grow low quality wood;

this is quite simply untrue. The variable Welsh environment produces a wide range of timber types

but we do not select out and grade the more valuable softwood types effectively at the moment.

Sitka spruce has become the main species of choice for the modern UK sawmilling sector and

homegrown softwood now makes up 40% of UK consumption (Llewellin N. 2014, 8). Furthermore in

2011 the UK produced nearly 3.3 million m3 of sawnwood; Latvia produced 3.4 million m3 (FAO

2013). Britain is no longer a minor timber producer and has already clearly demonstrated that

homegrown Sitka spruce is fit for purpose in a competitive, demanding market dominated by well-

established foreign suppliers. Furthermore studies carried out by John Moore confirm that spruce

wood quality is improved by increasing rotation lengths. There may be no premium for growing

spruce with greater yields of higher stiffness/density wood at present (Moore J. et al 2011) but

because older stands already exist there is an opportunity to process and position higher grade

spruce within existing joinery or other higher value added markets. Older large diameter sawlogs

from stands in more sheltered Welsh sites could yield wide boards that are not available from

Scandinavian or Baltic sources. Furthermore larger diameter sawlogs may allow easier separation of

juvenile corewood from the much more stable higher density mature heartwood (Moore J. et al

2011).

In Wales some innovative thinking in regard to use of homegrown softwoods has been done by small

processors. This may partly be the result of larger sawmills driving the price of Sitka spruce but for

many years it was also the result of lack of demand for minor conifer species by larger mills, giving

smaller processors opportunities to buy those less favoured species at lower cost. However, the

government’s Renewable Heat Initiative has driven up log prices across the spectrum of conifer

species and log sizes. Also some larger processors such as Pontrilas Timber in Herefordshire now

routinely process minor species in order to reduce reliance on Sitka spruce. Therefore smaller

processors will need to add more value to homegrown softwoods in order to increase their margins

as conifer sawlogs increase in value. Could there be new opportunities for specialist sawmills to

process ‘oversize’ large diameter or other selected logs such as old but supressed small diameter

logs for higher value-added markets?

22

Wood cell structure and microfibril angle (MFA) Wood cell structure may seem

to be remote from the everyday life of wood processors and users; however it is the microstructures

within wood cells which determine wood properties which then affect the processing and usability

of wood. Microfibril angle or MFA is one of the most important characteristics in wood formation;

high MFA significantly lowers wood stiffness and causes longitudinal shrinkage. High MFA in juvenile

wood, spiral grained wood and compression wood causes all of the undesirable effects which are

outlined in following sections of this report. Drawing 1 below based on the schema in the USDA

Wood Handbook (USDA 2010, 3–7) shows how MFA is indicated by the winding angle of microfibrils

in the wood cell S2 layer. In the drawing S1, S2, S3, are the secondary layers, P is the primary layer and

ML is the middle lamella, the area between cells.

The bordered pit shown on the walls of the wood cell drawing also fundamentally changes wood

characteristics; some species, for instance Douglas fir, larch and Sitka spruce have pits which close

off (like valves) irreversibly during the wood drying process. This is known as pit aspiration and it is

responsible for causing timbers to be refractory; i.e. fluids will not pass easily through cell walls. Pit

aspiration makes chemical treatment of Douglas fir and larch very difficult without incising to the

depth of treatment required. On the other hand pit aspiration significantly reduces absorption and

adsorption of water in refractory species; this characteristic is of course highly desirable in window

joinery where rewetting occurs regularly. C. P. Ackers mentions the use of Scottish grown Sitka

spruce for oars; the wood’s lightness, springiness and resistance to soaking up water make the

species a good choice for this application (Ackers C. P. 1956, 53). Pit aspiration does not occur in pine

species, this is why they easily absorb treatments such as creosote and can make good fencing

products. However, untreated pine timbers used in window frames will also easily soak up water

causing swelling so that sashes lock into window frames. Poor design detailing or breakdown of

surface coatings on pine windows inevitably leads to rewetting and if not remedied, early decay.

Wood microstructures profoundly influence wood properties and performance of wood products.

23

Drawing 1: Wood cell structure showing microfibril angle (MFA) on S2 layer

24

Wood density is generally considered to be the most important of all wood properties and the

most significant factor in determining end use; it is often associated with strength and stiffness

(Zobel B. J. and van Buijtenen J.P. 1989, 7). High density also strongly correlates with durability

(Zobel B. J. and van Buijtenen J.P. 1989, 272), however density and stiffness (MOE) do not

necessarily correlate for all softwood species; indeed the only physical characteristic of wood

capable of effecting large changes in stiffness is the microfibril angle in the S2 layer of the tracheid

cell wall (Tsehaye A. 1989, 14). When Sitka spruce is graded for structural purposes in the UK,

stiffness is the limiting factor (Macdonald E. and Lee S. 2013). In other words, although density of

any given softwood species is a good indicator of suitability for applications, it is not the only factor:

It is a rather gross measurement (Zobel B. J. and van Buijtenen J.P. 1989, 9).

Density and moisture movement are strongly correlated so the higher the density, the more

shrinkage occurs in drying (Haygreen J. G. and Bowyer J. L. 1996b, 170). Therefore in critical

applications where cycles of wetting and drying occur, high wood density does not necessarily

impart the required performance. Indeed, the traditional use of high density Baltic or Russian pine in

British sliding sash windows is a typical mismatch of wood type to application; in wet weather they

often swell, making opening difficult. For exterior cladding where low moisture movement is

desirable, low density softwoods are the most appropriate so western red cedar and Japanese cedar

both perform well. Image 8 below shows wide Japanese cedar low density wide boards used for

exterior cladding and scribed onto a rustic stone wall. These cedar boards are around 500mm deep

and have not distorted or cracked, they are fitted between Japanese cypress poles which have

helical cracks indicating spiral grain.

25

Image 8: 500mm deep Japanese cedar cladding boards on a building in Kyoto

26

Wood density range within annual rings is also an important factor in determining the characteristics

of any given softwood type. Furthermore even within a species there can be a considerable spread

of values, for instance US grown southern pines the earlywood density can vary by 35% and

latewood by 29%. Latewood to earlywood width ratio and abruptness of transition also affect wood

characteristics. Radiata pine (Pinus radiata) has even-textured annual rings with a low latewood to

earlywood density ratio of around 1.8:1 (Zobel B. J. and van Buijtenen J.P. 1989, 7). Therefore

although fast grown radiata wood from New Zealand can have very wide growth rings (sometimes

up to 25mm width), it machines well and is now commonly sold as finished joinery products such as

stair components. However, Douglas fir (Pseudotsuga menziesii) can have a latewood to earlywood

density ratio of up to 5:1, so fast grown material can be very course textured and difficult to finish

(Zobel B. J. and van Buijtenen J.P. 1989, 8).

The within ring density range of spruce is reported by J. M. Harris to be less than that of radiata

pine. Generally the density variation across a growth ring far exceeds the density variation between

trees (Tsehaye A. 1989, 18). Welsh homegrown spruce is often dismissed as ‘low quality’ and ‘low

density’ and therefore selection of high grade logs is seen as irrelevant, however if within-ring

density variation is less than that of radiata pine then there may be opportunities to sell machined

spruce joinery components into higher value added markets. Basic densities of earlywood and

latewood quoted in John Moore’s Wood properties and uses of Sitka spruce in Britain give a within-

ring density ratio of 1.7:1 (Moore J. 2011, 14); this suggests that good results could be achieved from

machining selected Sitka spruce for joinery applications. Fertilizers tend to depress wood density in

conifers, however with Japanese larch they also result in more uniformity of density across growth

rings making the wood more workable and easier to machine. Uniformity within growth rings is very

important for efficient utilisation of wood and evidence suggests that for many conifers, fertilization

is linked to wood uniformity (Zobel B. J. and van Buijtenen J.P. 1989, 231).

Growth rate and ring width Growth rate is often mistakenly correlated directly with

wood density or ‘quality’ although there is ample literature clearly pointing to the complexity and

interaction of factors involved; actually for several species growth rate is not the main factor (Zobel

B. J. and van Buijtenen J.P. 1989, 9). Indeed when rapid growth rate is induced by silvicultural

interventions, wood density (and moisture content) may be independent of growth rate (Zobel B. J.

and van Buijtenen J.P. 1989, 219). In Britain this topic has been a major topic of debate amongst

foresters since at least 1955 with the publication of an article called Growing Quality Softwoods in

the Journal of Forestry which was written to refute reports by J. M. Turnbull and W. E. Hiley of radial

density variations in pine. The authors carried out an excellent study of density variations in samples

of imported Picea abies (Aldridge F. and Hudson R. H. 1958) but made the error of assuming that

patterns of wood density variation in that species were a model for density variations in pine.

During the late 19th century wood scientists studying growth ring width believed that slowly grown,

narrow ringed softwoods were of superior quality to fast grown, wide ringed softwoods. Silvicultural

practices that followed patterns observed in northern latitude forests such as close initial spacing

were therefore suggested in order to reduce the size of what we now call the juvenile corewood

zone. Despite objections from foresters with experience of wide ringed softwoods from wide spaced

plantations in South Africa and Australia these old ideas persisted until the early 1950s when

research started to show that growth ring width was not a valid characteristic by which to evaluate

wood qualities. This research demonstrated that close initial spacing of trees reduced the diameter

27

of the juvenile corewood zone and its influence on wood properties (Larson et al 2001, 6) but does

not seem to have influenced forest management significantly. Indeed it is arguable that mature

heartwood laid down over a larger diameter juvenile corewood will yield material no less useful as

that laid down over a small diameter juvenile core; consistent growth ring width and overall yield are

more important (Larson et al 2001, 5). In the American southern pines not only is there no

correlation between between wood density and ring width, wood taken from similar aged rings

counted from the pith but differing in width tend to show similar wood densities. However,

comparisons between wide ringed juvenile wood from plantations and narrow ringed, close grown

old growth timber are illogical and prone to error (Larson et al 2001, 7).

Zobel notes that the general rule for fir, spruce and hemlock is to have a high density near to the

pith decreasing for a few rings then levelling off or increasing slightly to the bark. Larch and Douglas

fir are similar to hard pines with a low density at the pith, a rapid increase in the juvenile zone and

levelling off towards the bark (Zobel B. J. and van Buijtenen J.P. 1989, 107). Many observers have

made the error of comparing wide ringed low density wood of young trees to that of narrow ringed

high density wood of older trees and then misinterpreting these properties as effects of growth rate.

Fast growth rate and high wood density are anyway not incompatible; ring width alone is a poor

indicator of wood properties (Zobel B. J. and van Buijtenen J.P. 1989, 179–180). Growth ring width

provides no reliable evidence of wood quality whereas ring width uniformity or rate of change have

significant effects on mechanical properties of wood (Larson et al 2001, 4–5). Jonathan Bawcombe

notes that rate of growth has no significant impact on properties of Douglas fir but tree age is

significant (Bawcombe J. M. 2011, 18). This is clearly demonstrated by the glulam section in image 9

below taken from Forest Products and Wood Science (Haygreen J. G. and Bowyer J. L. 1996a, 203).

28

Image 9: DF glulam section, the fastest grown lamella has highest density (Haygreen & Bowyer)

29

The effect of growth rate on density is counter-intuitive. Although it is commonly assumed that fast

growth causes decrease in density this is not usually the case. Species, age, tree vitality, and location

of wood in the tree are more closely related to density than growth rate. Softwoods do not exhibit a

consistent relationship between wood density and growth rate (Haygreen J. G. and Bowyer J. L.

1996a, 199–201) and there is potential for combining fast growth and high wood density.

Both narrow and wide ringed juvenile wood can be low density as can the very narrow ringed wood

towards the outside of senescent trees (Zobel B. J. and van Buijtenen J.P. 1989, 179–180); modern

wood science echoes the forgotten observations of Hiley and Ackers expressed sixty years ago (Hiley

W. E. 1955, 162). Narrow ringed, lower density wood from the outer zones of older large Douglas fir

could make ideal low moisture movement material for moisture-critical applications such as sliding

sash windows. Small diameter old suppressed trees could be equally useful in this regard because

narrow growth rings reduce end-jointing problems such as grain break-out associated with high

within-ring density ratios. Large Douglas fir boxed-heart baulks already have a high market value at

the moment and arguably demonstrate the ideal format for selling juvenile wood within a desirable

product. Selection of narrow ringed falling boards from the outer log during milling is not only

feasible but economically viable if that timber is then marketed appropriately. Narrow ringed

Japanese larch has been reported as showing some relationship between rate of growth and wood

density with narrow ringed material being densest, however other studies show no relationship

between ring width and basic density (Zobel B. J. and van Buijtenen J.P. 1989, 171–172).

Larger timber merchants often emphasise the importance of slowly grown trees with narrow growth

rings (“Russian and Latvian Sawn Timber” 2015). If imported slowly grown softwoods can be sold

better into high value markets using this unique selling point then why not Welsh-grown narrow-

ringed timber?

If the appropriate high value trees can be grown, then high-pruning or underplanting and long term

retention of British grown Douglas fir could be justifiable despite high labour and opportunity costs.

Indeed, a few English estates such as Stourhead and Longleat have already demonstrated that older

Douglas fir forests can be viable (Geruwich M. 2015). British markets already exist for Douglas fir

products with a range of growth ring widths and intelligent selection of wood properties to match

the demands of particular products allows a wide range of ring widths to be accommodated within

joinery applications. For instance, high stiffness wide ringed Douglas fir falling boards from larger

trees could make ideal stair stringers and handrails whilst narrow ringed material could make turned

balusters and newels. Larch is almost visually indistinguishable from Douglas fir so narrow ringed

larch could also be used for balusters and newels alongside wide ringed Douglas fir stringers.

Interestingly, if selected British-grown Sitka spruce logs can be shown to have even-textured growth

rings and low wood density variation within growth rings (as suggested in the literature) then large

diameter logs may have even more potential for joinery than homegrown Douglas fir. If this is the

case, older oversize spruce logs could become desirable specialist sawlogs.

30

Juvenile wood Juvenile wood is perhaps one of the main causes of wood variation in conifers

as well as of one the main causes of concern when considering the management and utilisation of

conifers. It is also one of the main causes of confusion in regard to the perceived quality of fast

grown, wide ringed wood. The larger diameter section of 15 year old Japanese larch to the right of

image 2 on page 7 above is all juvenile wood. It has growth rings up to 15mm wide, is of low density,

is ‘fluffy’ in character and has very narrow latewood rings. It would be erroneous to simply to

correlate its properties with those of older, mature fast grown wood or assume that all fast grown

wood is ‘weak’ or low density; however some researchers and commentators have done exactly that

(Zobel B. J. and van Buijtenen J.P. 1989, 82–83,87).

Juvenile wood is the first-formed wood near to the centre of the tree and can be clearly seen in the

47 year old Japanese larch section to the left of image 2 where it extends across five growth rings to

about 22mm width. The juvenile core of the 47 year old suppressed tree is hardly wider than one

ring width of the 15 year old open-grown larch and seems to support the notion that close spacing of

trees can suppress the growth of juvenile wood. However Zobel throws doubt on the economic

viability of close spacing (Zobel B. J. and van Buijtenen J.P. 1989, 94) although it does have currency

with enthusiasts of continuous cover forestry. Site class and stand density play significant roles in

determining the width of juvenile wood in Japanese larch (Zobel B. J. and van Buijtenen J.P. 1989,

83). The transition from juvenile wood typically occurs over several years and depending on species

and site conditions can take up to twenty years. Drawing 2 below shows the position of juvenile

wood in the stem of a tree and its effects on wood properties.

Drawing 2: Juvenile wood in a tree & its effects on properties (Kretschmann D. E. 1998)

31

The properties of juvenile wood are usually described thus; low density, a high percentage of

reaction wood, larger microfibril angles (MFA), short cell lengths, higher lignin and lower cellulose

percentages (Zobel B. J. and van Buijtenen J.P. 1989, 93). MFA is particularly significant and refers to

the helical winding angle of microfibrils in the S2 layer of the cell wall, the larger the angle the higher

axial shrinkage on drying. Therefore when juvenile wood is included in cut timber it will shrink more

than mature heartwood and cause either crook or bowing. High MFA, spiral grain and reaction wood

normally associated with juvenile corewood are less pronounced in Douglas fir (“Douglas-

Fir_stability.pdf” 2015). This makes drying of Douglas fir more straightforward and reinforces its

reputation for stability. According to a 1929 USDA study, Western hemlock is reputed to suffer less

spiral grain than Douglas fir (or the spruces) (Johnson R. P. A. 1929, 25). However on the other hand

reaction wood and spiral grain strongly influence the behaviour of juvenile wood in larch. The topic

has been of interest in Japan especially in regard to larch glulam manufacture because juvenile

corewood included in larch centre boards will often cause severe twist or ‘wind’ sometimes

alongside bowing and crook (Koga S. et al 2004). However, Japanese researchers at Sumitomo

Forestry have been researching the stability of glulam manufactured using three metre long larch

lamellae twisted up to 10mm out of true and have obtained good results. Using two twisted lamellae

within glulam made up out of a total of four lamellae, finished glulam components did not twist any

more than 3mm after 36 months (Ishigaki H. 2014).

Longer rotations have been suggested by some foresters in order to make the cutting out of juvenile

wood more viable. Then mature wood with its more desirable wood properties and higher stability

might be sold separately. This is certainly possible already in Wales; there are stands of large

Douglas fir and other minor species growing here and because some private forest owners have for

some years been holding back from harvesting, there is thought to be a ‘bulge’ of oversize Sitka

sawlogs available in the future. Japanese researchers recommend long term retention of Japanese

larch to around 100 years old so that high grade material can be separated from juvenile corewood.

High grade logs over 40cm diameter are deemed to be suitable for furniture (Hirokazu K., Masatoshi

Y., and Syuhei I. 1990, 1).

The contribution of B. H. Paul (often quoted by Zobel) to the study of juvenile wood has been

summarised thus (Larson et al 2001, 3):

Paul challenged the concept of juvenile wood. He argued that juvenile wood, or corewood, was not necessarily inherent to tree growth because wood characteristics of these centrally located growth rings were under silvicultural control. His argument was based on the fact that corewood did not contribute to either longitudinal shrinkage or distortion in the southern pines when the growth rings were narrow (1/8 to 1/10 in. (3.2 to 2.5 mm)). On the contrary, both structural defects were excessive when growth rings in the core wood were wide (>1/4 in. (6.4 mm)). Paul’s argument is tempting, and it is supported by numerous studies. However, as proven by so many other studies, slow growth does not eliminate juvenile wood but simply confines it to a smaller core. Both narrow- and wide-core wood rings retain their juvenile characteristics compared with narrow and wide mature rings, respectively. Paul, however, did correctly interpret the effects of early stand competition on the reduction of springwood formation, hence the increase in summerwood percentage in the juvenile growth rings. Early stand competition reduces crown size and encourages rapid upward retreat of the live crown;

32

both of which restrict springwood formation. This silvicultural option is open to forest managers if they are willing to sacrifice volume production for the more favourable small diameter of core-wood growth rings.

Rather than using silviculture to reduce diameter of corewood during early growth, juvenile wood

could be separated during conversion of sawlogs. Then a problematical material is produced

because of the properties outlined above; the propensity of juvenile wood for axial movement, low

stiffness and general instability rule out its use in structural applications. However, some

applications need only short sections of wood, for instance end-grain flooring and end-grain

worktops and there may be some scope for use of juvenile wood in other specialist applications. For

instance no studies have been undertaken in order to utilise axial movement in appropriate

applications. Unpredictable materials need a creative approach; twisted timber could be an

attractive decorative feature when used appropriately, design according to wood properties and

marketing are the keys to realising the potential of different wood types. As plantation forests

increase in importance for wood supply on an increasingly crowded planet, fast-grown trees which

are logged early will become the norm. Proportions of juvenile wood in sawlogs are likely to increase

with expansion of fast grown forest plantations and protocols for its use will have to be devised

(Plomion C., Leprovost G., and Stokes A. 2001)

Mature heartwood The concept of mature heartwood is not generally understood by wood

users and not fully understood by researchers. Haygreen and Bowyer record that Hillis reported an

average age of 14-18 years for onset of transformation of juvenile wood to mature heartwood

(Haygreen J. G. and Bowyer J. L. 1996a, 30). Generalisations about any mature heartwood cannot

hold true because so many wood properties vary radially from the juvenile transition zone outwards

to the sapwood; however Haygreen and Bowyer record the following properties of mature

heartwood (Haygreen J. G. and Bowyer J. L. 1996a, 33):

1 May be darker than sapwood

2 May be decay and insect resistant

3 Difficult to penetrate with liquids

4 May have a distinct odour

5 Low hygroscopicity

Radial trends vary greatly and some have more significance than others, for instance in Japanese

larch durability increases more or less linearly from pith to sapwood boundary and then abruptly

decreases (Dauksta D. 2011, 6). In both larch and Douglas fir wood density is low at the tree centre,

rapidly increases through the juvenile wood and then gradually levels off towards the bark (Zobel B.

J. and van Buijtenen J.P. 1989, 107). Also in both western larch (54%) and Douglas fir (37%) mature

heartwoods have relatively low moisture content compared to their sapwoods (USDA 2010, 4–2).

Normal heartwood generally appears to be more stable than juvenile wood and shrinks very little

along its length; around 0.1 to 0.2% (USDA 2010, 4–5). In some species the heartwood colour trends

may reflect changes in extractive content and durability. However heartwood colour is variable and

33

assertions are better avoided (USDA 2010, 3–13). Some species have heartwood that is visually

hardly discernible from the sapwood (Haygreen J. G. and Bowyer J. L. 1996a, 33).

Based upon the scientific literature and empirical observations of sawmillers, whatever radial trends

might be present in mature heartwoods it is arguable that in order to increase stability and produce

the highest grades of softwoods, juvenile corewood will generally need to be removed from mature

heartwood for use in high performance or moisture critical products.

Spiral grain occurs when wood fibres follow helical paths around the longitudinal axis of a tree

and it is considered to be one of the most serious of defects encountered in sawn timber. It tends to

be most severe towards the centre of the tree (Zobel B. J. and van Buijtenen J.P. 1989, 23) because

there is a greater tendency for spiral grain within the juvenile heartwood which also tends to exhibit

greater microfibril angles (MFA) in the S2 layer of the cell wall. High MFA can cause higher than

normal longitudinal shrinkage (Haygreen J. G. and Bowyer J. L. 1996a, 103). Therefore as it dries, in

sawnwood containing both juvenile and mature heartwood, the juvenile wood tends to shrink more

causing the board to bend. Spiral grain will on drying tend to cause centre boards to ‘wind’ or twist

(Fonwenban J. et al 2013, 1). Therefore scantlings and boards cut with the pith along one edge and

mature heartwood on the other may bow and twist, rendering the timber difficult to utilise. In some

softwoods spiral grain tends to increasingly wind to the left, this is called S-spiral, reaching a

maximum after 10 to 20 years. Then in the mature heartwood the grain straightens up and the spiral

increasingly winds to the right, this is called Z-spiral (Wilson K. and White D. J. B. 1986, 178).

Traditional carpenters in Japan say of larch that it spirals to the left for fifty years and then spirals to

the right for fifty years and so it should not be cut until it is over a century old (Eiji U. 2014). Actually

Japanese researchers have found that larch mature wood grain generally tends to spiral only a little

and in the opposite direction to that of the juvenile corewood and that degree of twist in mature

wood is highly correlated with growth ring number from the pith (Koga S. et al 2004). Spiral grain is

under considerable genetic control but its expression may be dependent on the local environment.

Studies show inconsistent results; nevertheless spiral grain may well be an adaptive strategy for

withstanding extreme loads imposed by snow or wind. For Sitka spruce a recent study concludes

that planting on exposed sites with prevailing winds, wide initial spacing or later heavy thinning can

all increase grain angle. This suggests that in order to grow high grade timber exposed sites should

be avoided (Fonwenban J. et al 2013, 2).

Image 10 below shows a three metre long 50mm thick larch centreboard twisted 20o out of true by

spiral grain within the juvenile corewood indicated by the green circle. In order to avoid degrade in

drying stacks there is little doubt that thicker larch centreboards (over 30mm thick?) with spiral grain

should not be included amongst higher grade material cut from heartwood away from the juvenile

zone. Larch centreboards could perhaps be dried separately for particular applications where their

properties are less problematical and perhaps even desirable. Spiral grain causes twisting of posts in

service (Fonwenban J. et al 2013, 1) and is often indicated in roundwood by appearance of helical

cracking extending along logs. Image 11 below shows a huge twisted Japanese cypress column

within the Todai-ji complex at Nara, Japan.

34

Image 10: Spiral grain has twisted this 3m*150mm* 50mm larch centreboard through 20o

35

Image 11: Spiral grain shown by a massive helical shake extending up this column at Nara

36

Reaction Wood Trees respond to imposed loads such as prevailing wind against stems or snow

on branches by forming reaction wood. Sloping ground, changes in light availability caused by

windthrow or interventions will also cause reaction wood to be formed to enable trees to move

reactively in order to optimise their positions for receiving light. In conifers this wood is called

compression wood. Along with spiral grain compression wood is the cause of significant timber

degrade, for instance through distortion on drying. Compression wood is heavily lignified, denser

(although weaker) and sometimes darker in appearance than normal wood. The microfibril angle in

the S2 cell wall layer can be as high as 45o therefore compression wood, unlike normal wood, will

shrink axially (Plomion C., Leprovost G., and Stokes A. 2001); average values for normal wood being

between 0.1 to 0.2% (USDA 2010, 4–5). Conifers reacting against imposed side loads such as

prevailing winds form compression wood on the leeward sides of their stems as in image 12 below.

Compression wood when included in boards and dimensioned timbers with normal wood will shrink

axially (as will juvenile wood) causing crook or bowing (Forest Research 2001). Larch is well known

for forming swept butts and for its stem sinuosity; affected stems will contain compression wood

and can be very difficult to convert and dry without distortion. Timber with compression wood

cannot be used in critical applications although there may be creative uses for swept or sinuous

boards for decorative applications. Creative application combined with correct marketing may even

generate higher added value from the unusual forms which compression wood can cause.

Image 12: Cross-section of a larch stem; compression wood formed to the right of juvenile core

37

Drying homegrown softwoods Drawing 3 below shows the shrinkage or distortion

patterns of timbers cut from normal wood (USDA 2010, 4–5). Reaction wood and spiral grain can

significantly add to these drying distortions. T shows cupping on a tangential board, R1 and R2 show

how radially sawn boards and timbers shrink with little distortion, R3 shows how radially sawn

timber shrinks more along growth rings. C, the centre board contains a high proportion of juvenile

corewood, not only will it shrink as shown but it may also twist when spiral grain is present.

Drawing 3: Characteristic shrinkage and distortion of normal wood (USDA 2010, 4–5).

High yield plantation-grown softwoods offer a range of challenges firstly in drying and then later in

utilisation. Despite 150 years of study our understanding of typical radial patterns in the cross

sections of conifer stems has been described as minimal. Even the terms used to describe the

varying types of wood found within the stem of conifers have been debated (Meinzer et al 2011)

although the generalisations ‘juvenile’ and ‘mature’ heartwood are commonly used by foresters and

wood processors without necessarily understanding the complexities of typical radial patterns from

pith to bark. All of the different features discussed in previous sections above such as juvenile wood,

spiral grain and compression wood can occur and change at different rates across a transition zone

between the pith and mature zones.

Mature heartwood suffers little longitudinal shrinkage during drying (except when reaction wood is

present). However, when both juvenile and mature heartwood occur in the same board, the juvenile

wood will shrink more than the mature wood with consequent dimensional change causing

problems such as bowing. Spiral grain in juvenile wood can cause boards to twist during drying;

boards cut from the centre of a sawlog (and containing a juvenile core) are particularly prone to

38

twisting. Juvenile stems may take on helical or sinusoidal forms which are hidden beneath layers of

mature heartwood until revealed during sawing; then the changing properties caused by the juvenile

heart ‘wriggling’ along a board’s length may cause various complex dimensional changes leading to

problems in machining and/or significant loss in yield. Image 13 below shows a board cut from a

larch tree; the sinuous form of the juvenile tree has dominated the form of the mature tree.

Nevertheless this board was successfully dried to 12% MC by following a simple protocol.

Image 13: A board cut and dried from a sinuous small diameter larch log

39

In attempting to dry sawn softwoods to make them useable in construction the changes in wood

anatomy across tree stems are only part of the range of challenges facing processors; moisture

content also changes from pith to bark. For example in Sitka spruce heartwood moisture content

may vary from 40% to 80%. However in the sapwood (the outer zone of wood closest to the bark),

moisture content in excess of 120% is encountered and values close 300% have even been found

(Moore J. 2011). Thus material properties can vary widely in one board. Modern sawmills convert

material at such high rates there is little scope to change sawing patterns in order to separate

juvenile heartwood from mature heartwood although the outer ‘falling’ boards with relatively high

stiffness and high moisture content are normally separated but sold into low value markets.

There may be potential for new scanning and selection procedures capable of classifying boards

according to density, end grain imaging, distortion types (e.g. cupping or bow), stiffness and

moisture content (“Microtec” 2015). However, at present sawmill timber selection and binning

infrastructures in Britain may work too slowly for such complex grading routines (Brownlie A. 2013).

Researchers have suggested that boards be selected and grouped according to moisture content

before kilning but in practice this does not happen generally. Therefore kiln charges may be

composed of boards with many different material properties and a wide range of moisture contents.

Risk aversion significantly influences kiln management as kilns grow in volume to accommodate the

huge increases in sawmill production rates and so final mean moisture content has to be kept fairly

high (around 18%) in order to take account of the varying timber types and distribution of final

moisture contents found across a whole kiln charge.

Cutting patterns can seriously influence drying distortion of timber (SIRT 2014). Across much of

Europe from France to the Baltic region, sawmills have traditionally relied on sawing hardwoods en

boule or ‘through and through’ whereby logs are broken down by making parallel cuts across the

transverse end face and down the length of logs. For drying, stickers are then placed between the

resulting full width double waney-edged boards to give the appearance of a ‘reassembled’ log. This

method of drying is still standard practice for hardwoods across the world. Image 14 below shows

Welsh-grown Douglas fir logs which have been sawn, stacked en boule and successfully air dried.

Even relatively large sawmills in southern Germany sometimes air dry some of their softwoods in

this way and it may be one practical solution to the problem of drying British conifer timbers with

their widely varying radial properties. Although the three centre Douglas fir boards in image 14

contain both juvenile and mature zones, the juvenile core is bound within mature zones along both

edges thus balancing drying stresses and reducing distortion. Anyway in Douglas fir spiral grain does

not dominate behaviour of the juvenile core (“Douglas-Fir_stability.pdf” 2015) therefore

centreboards may be left within drying stacks. However, spiral grain is significant in larch juvenile

wood and to reduce risk to the whole stack from distorted centreboards, it is better to remove them

from stacks destined for high quality products. Double waney-edged boards can be processed

through double band resaws or multirip saws for final dimensioning by taking off both waney edges

simultaneously. This optimises width of each board and mature heartwood either side of juvenile

material constrains the behaviour of the corewood. This traditional method has been studied in

some depth and modified by American and Asian researchers with a view to obtaining better

conversion yields from difficult hardwoods and fast grown softwoods; it is called Saw-Dry-Rip or SDR.

40

Image 14: High grade Welsh grown Douglas fir cut ‘through & through’ then successfully air dried

SDR or en boule methods are unlikely to be taken up by high volume softwood sawmills in Britain for

producing joinery or other high grade timber but may appeal to smaller processors who wish to

differentiate their products and sell high grade softwoods into niche markets. When building drying

stacks there may be scope to select out centre boards which include pith and much of the juvenile

heartwood; these are the boards that are most likely to twist and induce distortion within stacks.

Large drying stacks of softwood do not necessarily need to be assembled en boule; actually

randomly distributed double waney-edged boards may dry more successfully within a stack which is

randomly distributing drying stresses; this needs more study. The most important factor is that

boards are not resawn whilst ‘green’ in a cutting pattern which encourages distortion when different

wood types interact asymmetrically such as when juvenile corewood is included on only one side of

a board. Image 15 below shows a stack of 30mm thick larch sawn ‘through and through’ and kiln

dried (under restraint) en boule, this timber has remained straight and is ready for edging and

dimensioning.

41

Image 15: 30mm thick larch sawn ‘through and through’ and then kiln dried under restraint

42

Bespoke sawmillers may have some advantage over volume producers; they often use horizontal

bandsaws which by their design allow through and through cutting. Traditional en boule drying of

softwoods may offer relatively easy value adding opportunities for small sawmills seeking specialist

markets. However there is at least one medium sized Welsh sawmill using the ‘MEM Teletwin’ saw

which because of its between centres design is ideal for sawing full width boards from either side of

the juvenile core. This topic is worthy of more study especially as softwood sawmilling becomes

more polarised between high volume and bespoke processors. Drawing 4 below shows a pre-edged,

centred-cant sawing pattern which also produces boards with roughly symmetrical properties which

have the best chance of drying with low degrade. The MEM Teletwin can use this cutting pattern.

More information about the MEM Teletwin here; http://www.memwood.com/gb/teletwin.html and

here; https://www.youtube.com/watch?v=ryLGKcZq2Tg

Drawing 4: Centred cant sawing pattern for SDR method, centre circle indicates juvenile zone

43

Summary Consistency of form and material characteristics within softwoods is highly desirable

but as demonstrated by studies of wood variation, difficult to achieve in practice. If there is one

message which emerges repeatedly in research literature then it is that wood properties vary more

within one tree or one forest stand than between forest sites. Different authors agree that exposed

sites may increase risk of trees forming reaction wood, spiral grain and decreased stiffness although

the picture is by no means clear as to how all the contributory factors interact to cause wood

variations. However, several commentators agree that planting in sheltered sites, avoiding wide

spacing, heavy thinning and other traumatic changes all contribute to growing high grade softwoods.

Sheltered parts of Wales are amongst the best places in Britain for growing conifers and Welsh

forests already grow some of the largest conifers in Europe. Western Britain is capable of growing

high grade softwoods such as Douglas fir comparable to plantation timber grown anywhere in the

world. Furthermore Britain now produces as much sawn softwood as Latvia; we are no longer minor

producers but our thinking in regard to our conifer forests has not matured accordingly.

It is still a commonly held belief that fast growth of conifer species in Britain produces low grade

timber; this is a biased, over simplified normative judgement. Growth rate or ring width is less

important for machinability than within growth ring density variation. Growth rate does not always

correlate directly with average wood density or wood stiffness; consistency of growth rate is more

important than growth rate considered alone for production of high grade timber.

Mistakes were made in New Zealand during establishment of plantation forests in the 1920s and

1930s; they learnt from their mistakes and used science to understand the resource and create a

very successful forest industry which contributes around 3% of GDP. British foresters in the 1950s

were accused of growing ‘rubbish’ because of the fast growth of conifers in the UK; this attitude was

based on flawed research. Wood qualities need to be defined in terms of end use and fitness for

purpose otherwise wood quality becomes an arbitrary term.

Anti-conifer forest stances and drip feed of negativity from NGOs and well-known journalists have

skewed political and public perceptions of planted forests; leading researchers regard this negativity

as ‘misinformation’.

Planted forests are now considered to be foundations for creating sustainable economies by the

Food and Agriculture Organisation of the United Nations (FAO) and the World Wildlife Fund (WWF).

However planted forests are exposed to socio-economic risks due to governance failures; these risks

comprise a weak or inadequate forest policy framework including insecure investment conditions

and current trends in European forest management could result in a shortfall of coniferous timber

(FAO).

Foresters, processors and researchers need to recognise the variability of wood, learn to live with it

and learn how manipulate it; wood variation allows choice of particular properties and selling points

from a range of grades. Softwoods are among the most versatile of natural products and in Britain

we grow softwoods with a very wide range of properties. We have not fully explored the use of

those different properties. Wood density is highly variable and there are value added applications

across the range of wood density. For instance where low moisture movement is desirable, low

density softwoods can be the most appropriate.

44

British forest plantations are generally not retained after year 50, before ‘old growth’ characteristics

start to develop, normally over 65 years old. This is when self-pruning occurs and trees grow clear

wood over knotty cores and forests take on a more mature appearance. Growing Douglas fir with

western hemlock or spruce in the understorey imitates old growth forests of the Pacific northwest

coast of America and can reduce branchiness of Douglas fir.

Scottish and Welsh researchers have demonstrated that British softwoods can be used in a range of

contemporary construction techniques. However technical feasibility for construction does not

guarantee economic viability, we need to understand how to make more profit from our softwoods.

Species such as Douglas fir (already valuable on global markets) offer opportunities for

differentiation and selling into high value-added markets where economic viability is not so

dependent on the strength of the pound against other currencies. Old growth Douglas fir is no

longer available from America; English merchants are selling British Douglas fir alongside American

second growth/plantation Douglas fir at up to £1000 per cubic metre, the same price as joinery oak.

Most large, valuable, older Douglas fir sawlogs grown in Wales have been sold to specialist sawmills

in England thus losing potential local economic gain in Wales and reinforcing negative views about

opportunity cost of growing big timber. Western hemlock, useful as an understorey species, was

considered to be a high value joinery wood historically but its value has been forgotten recently.

Spiral grain and reaction wood normally associated with juvenile corewood are less pronounced in

Douglas fir. This makes drying of Douglas fir more straightforward and reinforces its reputation for

stability. Western hemlock is reputed to suffer less spiral grain than Douglas fir and selected logs

could produce high grade timber. However on the other hand reaction wood and spiral grain

strongly influence the behaviour of juvenile wood in larch making the juvenile core unstable; it may

be necessary to cut out juvenile corewood in order to produce higher grades of larch timber.

The Japanese have built some of the largest timber structures in the world with their softwoods; the

British are (with a few honourable exceptions) still making fences with the same wood species.

Although we have started using homegrown C16 softwoods in house construction, the British have

been very conservative in their approach when compared with the structures they have made using

imported softwoods.

The Welsh environment has distinct, positive attributes for growing high grade softwoods; e.g. mild

wet autumns allow growth of denser, stronger latewood zones within growth rings. Selected

plantations could be subject to long term retention in order to study stem improvement with age

and to supply large diameter high grade logs; selected conifer species on sheltered Welsh sites will

grow valuable softwoods.

There are simple protocols called ‘Saw-Dry-Rip’ used in America and Asia for processing and drying

difficult timbers; here in Wales SDR or the related European method of drying ‘en boule’ can

produce stable, high grade softwoods. This is a simple step for SMEs, needing little investment and

capable of producing homegrown softwoods for high value applications.

45

Appropriate design according to wood properties and then marketing appropriately are the keys to

realising the potential of different wood types grown in Wales. More collaboration is needed

between designers and the timber industry. Wood is attractive, even beautiful; how often do we use

these words in connection with homegrown softwoods? The Japanese term wabi-sabi sums up their

respect for the aesthetics of natural processes, for the transcience and imperfection of natural

materials. Their traditional timber architecture and approach to design in any type of wood

demonstrates an understanding and respect from which we could benefit.

46

Bibliography Ackers C. P. 1947. Practical British Forestry. London: Oxford University Press. ———. 1956. “Whither Away?” Journal of Forestry L (1): 51. Aldridge F., and Hudson R. H. 1958. “Growing Quality Softwoods: Variation in Strength and Density

of Picea Abies- Specimens Taken from a Commercial Consignment.” Journal of Forestry LII (2).

———. n.d. “Growing Quality Softwoods: A Critical Examination of the Turnbull Hypothesis.” Journal of Forestry XLIX (4).

Bawcombe J. M. 2011. “South West Douglas Fir Timber Quality.” http://www.forestry.gov.uk/pdf/SIRT_workshop_15Nov2011_douglasfir_%28Bawcombe%29.pdf/$file/SIRT_workshop_15Nov2011_douglasfir_%28Bawcombe%29.pdf.

Bishop M. 2014. Planting new broadleaved forests. Brownlie A. 2013. Sawmill grading and binning. Capricorn Eco Timber. 2013. Value of homegrown Douglas fir. Dauksta D. 2011. Japanese Larch in Wales. Machynlleth: Woodknowledge Wales. ———. 2013. Welsh Softwoods in Construction. Machynlleth: WFBP.

http://www.wfbp.co.uk/files/WSW%20Welsh%20Soft%20Wood%2021%2011.pdf. Dearing Oliver et al. 2014. “Carbon, Fossil Fuel, and Biodiversity Mitigation With Wood and Forests.”

Journal of Sustainable Forestry 33 (3). http://www.tandfonline.com/doi/full/10.1080/10549811.2013.839386#abstract.

“Douglas-Fir_stability.pdf.” 2015. Accessed February 13. http://www.douglasfir.co.nz/_uploads/downloads/douglas-fir_stability.pdf.

Dykstra D. P., and Monserud M. A., eds. 2007. “Forest Growth and Timber Quality: Crown Models and Simulation Methods for Sustainable Forest Management.” In , 116, 139, 140, 170. Portland: USDA.

Edwards C. 2015. “20-30 Year Old Spruce.” Eiji U. 2014. Spiral grain in larch. FAO. 2013. Forest Products. 46. Forestry Series. 2013: FAO. Fonwenban J. et al. 2013. “Modelling the Effect of Spacing and Site Exposure on Spiral Grain Angle

on Sitka Spruce (Picea Sitchensis (Bong.) Carr.) in Northern Britain.” Forestry. Forest Research. 2001. Compression Wood Programme Summary.

http://www.forestry.gov.uk/fr/INFD-4XGJC8. Franklin J. F., and Spies T. A. 1983. “Characteristics of Old-Growth Forests.” In New Forests for a

Changing World : Proceedings of the 1983 Convention of the Society of American Foresters, Portland, Oregon, October 16-20. Portland: Society of American Foresters.

Gardiner B. 2015. “Wind Influence on Wood Properties.” Gartner B., Robbins J., and Newton M. 2005. “EFFECTS OF PRUNING ON WOOD DENSITY AND

TRACHEID LENGTH IN YOUNG DOUGLAS-FIR.” Wood and Fiber Science 37 (2): 304–13. Geruwich M. 2015. Large Douglas fir, WH understorey. “GLOBAL WIND MAP.” 2015. Accessed March 4.

https://maps.google.com/gallery/details?id=zJuaSgXp_WLc.kTBytKPmNODY&hl=en. Goldin I. 2014. Is the Planet Full?. Oxford: Oxford University Press. Haygreen J. G., and Bowyer J. L. 1996a. Forest Products and Wood Science an Introduction. Ames:

Iowa State University Press. ———. 1996b. Forest Products and Wood Science. Ames: Iowa State University Press. “Henry John Elwes.” 2015. Accessed February 11. http://www.colesbournegardens.org.uk/about-

us/henry-john-elwes.html. Hiley W. E. 1955. “Quality in Softwoods.” Journal of Forestry XLIX (3). Hirokazu K., Masatoshi Y., and Syuhei I. 1990. “Experiments for Utilizing Large-Diameter Larch Logs

Grown in Hokkaido.” Journal of the Hokkaido Forest Products Research Institute 4 (6). Ishigaki H. 2014. Stability of glulam using twisted larch lamellae.

47

Johnson R. P. A. 1929. Properties of Western Hemlock and Their Relation to Uses of the Wood. Technical Bulletin. Madison: USDA. http://naldc.nal.usda.gov/download/CAT86200133/PDF.

Koga S. et al. 2004. “Warp in Dimension Lumbers of Japanese Larch.” In . Kretschmann D. E. 1998. “Properties of Juvenile Wood.” USDA.

http://www.fpl.fs.fed.us/documnts/techline/properties-of-juvenile-wood.pdf. Larsen P. R. 1969. Wood Formation and the Concept of Quality. Bulletin 74. New Haven: Yale

University: School of Forestry. Larson et al. 2001. Formation and Properties of Juvenile Wood in Southern Pines A Synopsis. General

Technical Report FPL − GTR–129. Madison: USDA. http://www.fpl.fs.fed.us/documnts/fplgtr/fplgtr129.pdf.

Lavers G. M. 1969. The Strength Properties of Timbers. 2nd ed. London: HMSO. Liang J. et al. 2009. “Optimizing Wood Quality and Stand Diversity in Uneven-Aged Forest

Management.” In Forest Growth and Timber Quality: Crown Models and Simulation Methods for Sustainable Forest Management, 173. Portland: USDA.

Llewellin N. 2014. Statistical Review 2014 INDUSTRY FACTS AND FIGURES FOR THE YEAR 2013. Timber Trade Federation. http://www.ttf.co.uk/timber-industry/statistics.aspx.

Macdonald E., Gardiner B., Mochan S., Fonweban J., and Moore J. 2008. “Trees for the Future.” In . Cumbernauld.

Macdonald E., and Lee S. 2013. “Volume or Quality?” presented at the SIRT Timber research Workshop. http://www.forestry.gov.uk/pdf/Volume_or_quality%28ElspethMacdonald%29.pdf/$file/Volume_or_quality%28ElspethMacdonald%29.pdf.

Macdonald M., and Gardiner B. 2005. The Implications of Transformation to Continuous Cover Forestry Systems for Conifer Timber Quality and Log Supply in the UK. Roslyn: Forest Research. http://www.forestry.gov.uk/pdf/Implications_of_transformation_to_CCF_conifer_timber_quality.pdf/$file/Implications_of_transformation_to_CCF_conifer_timber_quality.pdf.

Mattheck C. 1998. Design in Nature: Learning from Trees. Berlin: Springer-Verlag. McArdle R. E., and Meyer W. H. 1949. The Yield of Douglas Fir in the Pacific Northwest. Technical

Bulletin 201. Washington, DC: USDA. McCarthy M. 2010. “Nature Studies by Michael McCarthy: Never Has the Deer Been so Dear to Us

Nature Studies Friday 30 April 2010.” The Independent, April 30. Mclean P. 2014. Softwood properties/qualities in Britain. Meinzer et al. 2011. Size- and Age-Related Changes in Tree Structure and Function. Springer. Mendes A. et al. 2013. “Planted Forests Are a Vital Resource for Future Green Economies.” In , 6–13.

FAO. http://www.fao.org/forestry/37902-083cc16479b4b28d8d4873338b79bef41.pdf. Met Office. 2014. UK Wind Map. Exeter: Met Office.

http://www.metoffice.gov.uk/media/pdf/l/7/14_0058_Site_search_for_sml_med_wind_projects.pdf.

“Microtec.” 2015. Accessed February 23. http://www.microtec.eu/. Moore J. 2008. “Understanding the Variability in the Wood Properties of Home-Grown Timber.”

http://www.forestry.gov.uk/pdf/exeter0408_moore.pdf/$file/exeter0408_moore.pdf. ———. 2011. Wood Properties and Uses of Sitka Spruce in Britain. Forestry Commission.

http://www.forestry.gov.uk/pdf/FCRP015.pdf/$file/FCRP015.pdf. Moore J. et al. 2009. “The Effects ofSite and Stand Factors on the Tree and Wood Quality of Sitka

Spruce Growing in Britain.” Silva Fennica 43 (3). ———. 2011. “Effects of Rotation Length on the Grade Recovery and Wood Properties of Sitka

Spruce Structural Timber Grown in Great Britain.” Annals of Forest Science. NBS. 2014. “NBS | National Building Specification | Timber/joinery.” Accessed December 15.

http://www.thenbs.com/topics/ConstructionProducts/bibliography/timberJoinery.asp.

48

NZFOA. 2013. New Zealand Plantation Forest Industrry Facts & Figures. Wellington: New Zealand Forest Owners Association. http://nzfoa.org.nz/images/stories/pdfs/foa_facts_figures_2013.pdf.

Plomion C., Leprovost G., and Stokes A. 2001. “Wood Formation in Trees.” Plant Physiology 127: 1513–23.

Prior G. 2015. “Timber Frame Specialist Goes into Administration.” Construction Enquirer. Accessed March 5. http://www.constructionenquirer.com/2015/03/02/timber-frame-specialist-goes-into-administration/.

Ridley-Ellis D. 2015. “Larch Grading Settings Update.” “Russian and Latvian Sawn Timber.” 2015. Accessed January 30.

http://www.finnlumber.com/index.php/9-main/40-russian-and-latvian-sawn-timber. Scott C. W. 1960. “Radiata Pine as an Exotic.” Unasylva 14 (1). Singleton R., DeBell D. S., Marshall D. D., and Gartner B. 2003. “Eccentricity and Fluting in Young–

Growth Western Hemlock in Oregon.” Western Journal of Applied Forestry 18 (4). http://www.fs.fed.us/pnw/olympia/silv/publications/opt/473_SingletonEtal2003.pdf.

SIRT. 2014. Impact Case Study: Mechanical Properties of UK Grown Timber - Improving Quality and Reducing Waste. Edinburgh: Edinburgh Napier University.

Taylor W. L. 1958. “How Should We Grow Conifers?” Journal of Forestry LII (1): 22–24. Telewski F. W., and Jaffe M. J. 1986. “Thigmomorphogenesis: Field and Laboratory Studies of Abies

Fraseri in Response to Wind or Mechanical Perturbation.” Physiol. Plant 66 (211-8). “This Is Money.” 2015. Accessed February 18. http://www.thisismoney.co.uk/money/news/article-

2877859/Pound-strengthens-against-42-currencies-year-spelling-better-news-tourists-heading-Scandinavian.html.

Tsehaye A. 1989. “Within and Between Tree Variations in the Wood Quality of Radiata Pine.” University of Canterbury, N. Z. http://www.google.co.uk/url?sa=t&rct=j&q=&esrc=s&source=web&cd=2&ved=0CCsQFjAB&url=http%3A%2F%2Fir.canterbury.ac.nz%2Fbitstream%2F10092%2F7147%2F1%2Ftsehaye_thesis.pdf&ei=jra7VPjVLJDnarbjgtAK&usg=AFQjCNGjbcq9mSsEeet464Wacl4i9aRJAw&bvm=bv.83829542,d.d2s.

Tsehaye, A. 2015. “Within- and between-Tree Variations in the Wood Quality of Radiata Pine.” Accessed January 30. http://ir.canterbury.ac.nz/handle/10092/7147.

TTF. 2014. TTF Statistical Review 2014. London: Timber Trade Federation. http://www.ttf.co.uk/timber-industry/statistics.aspx.

USDA. 2010. Wood Handbook Wood as an Engineeering Material. Madison: USDA. http://www.fpl.fs.fed.us/documnts/fplgtr/fpl_gtr190.pdf.

Waugh Thistleton Architects. 2015. “Murray Grove.” Accessed March 4. http://www.waughthistleton.com/project.php?name=murray.

Wilson K., and White D. J. B. 1986. The Anatomy of Wood: Its Diversity and Variability. London: Stobart and Son.

Woods R. 2015. Branchiness in conifer plantations. WWF. 2013. Forests and Wood Products. WWF. http://www.worldwildlife.org/publications/wwf-s-

living-forest-report-chapter-4-forests-and-wood-products. Zobel B. J. 1992. “Sivicultural Effects on Wood Properties” 2: 31–38. Zobel B. J., and van Buijtenen J.P. 1989. Wood Variation Its Causes and Control. Berlin: Springer-

Verlag.

49