The Future

Abstract

Much has been achieved in the domestication of radiata, it being among the most domesticated of all forest trees. A broad array of technologies has contributed in both growing radiata as an intensively managed crop and achieving genetic improvement, the two main planks of domestication. Refinements in the technology for processing and utilisation of the wood, however, are also part of the broader issues of domestication. Such refinements may obviate the need for improving certain wood properties by genetics or management inputs. On the other hand, the same refinements may help to identify further genetic improvements that need to be made. In this context, much of the further progress in domestication is likely to involve genetic improvement.

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

9.1 Introduction

Much has been achieved in the domestication of radiata, it being among the most domesticated of all forest trees. A broad array of technologies has contributed in both growing radiata as an intensively managed crop and achieving genetic improvement, the two main planks of domestication. Refinements in the technology for processing and utilisation of the wood, however, are also part of the broader issues of domestication. Such refinements may obviate the need for improving certain wood properties by genetics or management inputs. On the other hand, the same refinements may help to identify further genetic improvements that need to be made. In this context, much of the further progress in domestication is likely to involve genetic improvement.

Radiata plantations have also provided the context for many technological developments in plantation forestry. Those developments have been in fields that include management and decision-aid systems, harvesting technology and ergonomic research. Radiata will doubtless feature in further developments. That said, we do not address the full gamut of ramifications.

So far, breeding has served in part to mitigate the ill-effects of intensive management practices designed to both boost yields and otherwise improve profitability by reducing harvest ages. In the longer term, the significance of breeding for this role and for mitigating the impacts of other factors that detract from the yield or value of the crop will presumably increase. The identity of these factors will vary with location and site. They may include new pests and diseases, climate change, either intensive fertilizer use or nutrient depletion, and adverse physical changes in soils. That much forest plantation will doubtless occupy inherently poor soils is relevant; lapses in management can quickly squander the results of decades of careful husbandry (e.g. O’Hehir and Nambiar 2010).

Despite the future importance of continued genetic improvement, replenishment of nutrient resources will remain an issue. Fertilizer will often be needed to boost or maintain yields on finite land areas. Depletion of some nutrients can easily be made good by a combination of minimizing losses, geological weathering, and replacement with fertilizers. Nitrogen and phosphorus, however, remain problematic. Nitrogen fertilizer will tend to be expensive because of its high energy requirements. Significant N fixation (6–24 kg ha⁻¹ yr⁻¹) can occur in radiata plantations in the absence of leguminous ground vegetation (Turner and Lambert 2011), but this will depend on general soil fertility and adequate humus. Depending on leguminous vegetation to fix nitrogen will require light which in turn means less than full site occupation by the pines. Recent, unpublished work indicates that applying nitrogen to mid-rotation stands can be more cost-efficient than was believed, possibly with little downside for wood properties, while leaching of the nitrogen into waterways should not be important in such stands. Phosphorus fertiliser may become scarce in future (Vaccari 2009; Ashley et al. 2011), which is a worry because radiata is relatively phosphorus-demanding among pines. Yet the phosphorus demands for radiata crops are likely to be very minor compared with those for food crops and pasture. Moreover, genetic variation within radiata in phosphorus demands could presumably be exploited if need be. Furthermore, judicious use of phosphorus fertilizer for radiata should not lead to troublesome eutrophication in streams and lakes.

However, for both setting the agenda for genetic improvement and managing forest estates, prediction of crop performance is going to be very important. Such predictions will in turn depend on modeling growth and wood properties through to harvest age. Such modeling is admittedly well advanced, but there remains much scope for refinement, especially as the genetic makeup of plantations will be subject to continuing change.

Before considering what future developments in the domestication might be, it is appropriate to take more detailed stock of the present state of domestication. In important respects it is still very incomplete, which poses plenty of remaining challenges and opportunities. Let us, then, look at the key respects in which radiata is still essentially a wild organism.

9.2 Domestication Gaps and their Implications

Forest trees in general are genetically in essentially a wild state. With limited exceptions, their natural variability is essentially intact, and most species are resistant to management under the inbreeding regimes that are needed to produce genetic uniformity in various major crop plants. However, where clonal replication is an option, it can avoid unwanted variation arising from the genetic segregation that occurs during seed production. Provided an infrastructure of a breeding population is maintained and properly managed, it is possible to both retain genetic diversity and yet enjoy the benefits of crop uniformity. Moreover, the vast majority of forest trees are not even subject to plantation forestry, and few of them even look like promising candidates for the future.

Radiata, despite the considerable domestication that has been achieved, remains largely a wild organism in three general respects: the state of its genetic system, its reproductive biology and its physical architecture. We now review these three respects and their implications.

9.2.1 The State of the Genetic System

Radiata is still very much an outbreeder, dependent on cross pollination among essentially unrelated individuals, with much genetic variation among individuals, and high heterozygosity (effectively genetic variation between duplicate chromosomes within individuals). All this means that plenty of continued genetic gain can be expected from classical selective breeding. By the same token, the species is not amenable to close and prolonged inbreeding in order to ensure crop uniformity and/or to capture dramatic heterosis (hybrid vigour) as is widely practised in maize breeding. Repeated self-fertilisation can give widely varying outcomes among inbred lines of radiata (cf Wu et al. 2004 and preceding papers in series). In some cases it is thereby possible to “purge” highly deleterious genes (“hard genetic load”), after which the lines may regain self-fertility and vigour. In other cases, inbred lines may slide over time to extinction, a situation that is thought to reflect epistatic genetic load, through which continued inbreeding causes insidious loss of function among various alternative developmental pathways. The potential of intensive inbreeding as a practical breeding tool in radiata is thus very problematic. In any case, use of inbreeding is not needed for obtaining desired crop uniformity if it is possible to practise mass-propagation of clones.

Associated with the wild-state outbreeding is a virtual lack of linkage disequilibrium (LD) between genetic loci (Burdon and Wilcox 2011), except within families. This means that between sites (or loci) on individual chromosomes the co-occurrence of different alleles (alternative forms of genes) is essentially random. As a result, the use of selectively neutral genetic markers, which have no direct bearing on fitness or economic traits, for selecting superior individuals is not generally possible. Use of association genetics, relating particular DNA variants at particular loci to trait expression, is possible in principle. For that, purely empirical studies look unpromising, but use of candidate genes based on known roles in other plant species is more promising (Burdon and Wilcox 2011). The increasing practicality of genome-wide scanning, using very large numbers of DNA markers, is prompting the pursuit of genomic selection whereby very early selection can be done just on DNA information with a reliability that is acceptable in relation to the time saved. This is being done in centralised modern dairy-herd breeding (e.g. Hayes et al. 2009a; De Mello et al. 2014), where LD is much greater. For forest trees, with very limited LD, the population sizes for which genomic selection will work satisfactorily are still problematic, and may be smaller than what are strategically appropriate. For successful application of such genomic selection, the dramatic reduction in cost identifying DNA markers is crucial, but the huge size of the pine genome will surely be a complication.

9.2.2 Reproductive Biology

Both sexual reproduction (production of seed) and vegetative reproduction (classically, grafting or producing cuttings) are involved.

Seed production is essential for selective breeding, since it involves the genetic recombination that allows the breeder to produce ever better individuals to choose from and thus provide cumulative genetic gain. And for various breeding purposes, the earlier in the tree’s life seed can be produced, the better. On the other hand, producing pollen, conelets and thence seed typically consumes considerable resources. While hard to demonstrate rigorously, this surely represents a diversion from producing wood—to a degree, “making love, not wood.” Moreover, sexual reproduction, especially in the phase of pollen shedding and dispersal, is likely to breach containment of transgenes if genetic engineering is practised.

An ideal for long-term domestication is to have flowering (and thence pollen and seed production) “on command and command only.” Whether this is a realistic goal is controversial. Extremely precocious flowering has not been induced by using gibberellin on pines in the way that it can be induced in Cupressaceae. However, genes that can cause such flowering can evidently exist in Pinus (Smith and Konar 1969), although their amenability to management is quite unproven. Complete suppression of sexual reproductive structures in field-fit pines remains only a goal. Also, even if it is achieved, some adjustment of crown architecture may be needed (Burdon and Walter 2004), because the production of pollen in pines does influence crown architecture by creating zones on twigs that are bare of foliage. With elimination of pollen-producing structures these bare zones would disappear, which might mean that tree crowns become denser than is optimal.

Sexual propagation, using hybridisation, is the traditional means of combining desired attributes of different populations or species. We would certainly like to overcome the crossability barriers that currently allow hybridisation of radiata with very few other species, but that will almost certainly entail strong research commitment.

Vegetative propagation technology of radiata needs further refinement. With grafting, delayed incompatibility can be troublesome, and is not fully overcome, but it is no longer a major limiting factor. However, other forms of vegetative propagation, namely rooting of cuttings and in-vitro propagation technologies, are still beset with very important limitations. Young seedlings are easily multiplied as nursery cuttings, and this can be continued over a number of years with technologies that slow the process of maturation and allow continued propagation. Practising embryogenesis from seed tissues, in conjunction with cryopreservation using liquid nitrogen, allows repropagation of clones over long periods of time (Sect. 7.9.1). Embryogenesis also provides a platform for genetic engineering (Sect. 7.11.2). However, a restriction remains on the practice of clonal forestry in which radiata clones of choice can be mass-propagated on an unlimited scale over an indefinite time frame, because achieving this still depends on using fully-juvenile embryo material for cryopreservation. The ideal remains the ability to achieve reliable control over maturation whereby any individual tree, whatever the age, can be mass-propagated at will from whatever maturation state. In practice, this would entail being able to achieve full rejuvenation of any individual without incurring the genetic recombination associated with seed production.

9.2.3 Physical Architecture

In the wild, a tree needs to be able to get its crown where it can intercept enough light to compete and eventually reproduce itself. In the case of a light-demanding, pioneer species, like radiata, that means being able to compete vigorously with both other members of its own species and with other species. To achieve that, with a safety margin to avoid climatic damage (e.g. windthrow, or wind- or snow-breakage), highly efficient physical architecture is required. That architecture embraces not just the general form of the tree, but also patterns of within-tree variation in wood properties.

There are shade-tolerant tree species, whose competitive ability does not depend on rapid growth from an early age. Among conifers, these include many true firs (Abies spp.). Such species can, as stands get older, be extremely productive, but they are generally not amenable to the intensive domestication that has proved possible with radiata. When grown as plantations, their typical slow early growth delays effective site occupation so as to both erode per-annum wood production and delay harvest, both effects having serious implications for the economics of commercial forestry.

As competitors, forest trees typically have a highly efficient mechanical architecture, although the architectural strategies are very diverse among species. In pines, the radial and vertical gradients in wood properties, involving density, stiffness and fibre dimensions, can readily be explained as mechanical adaptations. These effectively mean that trees are achieving their competitive status by producing the minimum necessary amount of wood substance to present their crowns to the light while ensuring mechanical robustness. And the within-tree variations in wood properties, while they make for mechanical efficiency of intact stems, can be highly undesirable for wood processing and end-product performance; quality control can be badly affected, and solid-wood products can be dimensionally unstable during drying and in service.

It is instructive to digress into a development in crop plant breeding. With many crop plants, gains in yield have come mainly in harvest index (that is, the proportion of biomass that is of commercial value) rather than in primary biomass production (Evans 1993). A classic case is the dwarf wheats of the Green Revolution. The short stalks mean less diversion of resources into essentially unusable straw. They also mean that growth responses to fertiliser do not get vitiated by lodging which tends to happen with traditional long-strawed wheats, much to the detriment of grain yield. Herein is a particularly strong synergism between genetic improvement and management inputs. Yet the very features that make those dwarf wheats efficient grain producers means that they are poor competitors, acutely dependent on weed control.

So far, tree breeding practice has, in pursuit of increased growth, entailed selection for competitive ability. The high percentage of primary biomass that the bole represents in trees like radiata much reduces the scope for productivity gains through increasing harvest index. Nevertheless, there are almost certainly some productivity gains to be made through exploiting some divergences between competitive ability and productivity. Paradoxically, such gains could come from achieving mechanical inefficiency (or “overdesigning”) in trees. In such a scenario, wood can be produced in excess of mechanical requirements, and without the within-tree variations. The “crop ideotype” tree is suggested as having the following characteristics: a fat, cylindrical bole, with highly uniform wood, and far more strength and stiffness than is needed for mechanical stability; modest height after early stages of growth (cf Libby 1987; Tuskan 2007); a crown with light branches, possibly drooping, with renewal of crown not depending on continued height growth; and no diversion of resources into reproduction. Fast early growth, however, will still be needed, to achieve the rapid site occupation that is a prerequisite for high productivity on an economic time scale. And the righting mechanisms needed to recover from minor mechanical damage will still need to function. But even with these features the crop ideotype will be a poor competitor. That, however, should be no disadvantage if the trees are grown in monoclonal blocks, within which their competitive limitations would not matter. Pursuit of the theoretical crop ideotype, however, rests uncomfortably with the observation that, in largely undomesticated radiata, stem diameter is less favourably correlated with quality traits than is height (Burdon 1992, table 3), suggesting a stiff challenge in achieving truly intensive domestication.

The issue of mechanical architecture also intersects with manipulation of soil fertility. Experience with growing radiata on highly fertile ex-pasture sites is that productivity can be strongly boosted, with an increase in wood volume of 15−25% at harvest (Maclaren and Knowles 1999; Maclaren and West 2005). Unfortunately, tree form can be badly affected, with gross branching and much malformation, and wood properties can be very badly affected (Burdon 2010a; Hawke 2011). The gross branching makes the trees subject to top breakage. Moreover, while some flexibility of the stem and crown is desirable for reducing wind resistance, excessive flexure resulting from wood of lower stiffness is likely to make trees more vulnerable to the gustiness caused by the wind turbulence that occurs within the stand canopy. In short, the developmental responses that are adaptive under low to moderate soil fertility appear to become counterproductive under high soil fertility. Achieving adaptation to the higher fertility that can boost productivity, without adverse effects on physical architecture, will be a challenge for the future.

Radiata belongs with almost all other forest tree species in that physical architecture has not been addressed in connection with genetic improvement, except at a very crude level. Yet if the issue is to be addressed in depth, radiata must be very high among the list of forest tree species involved.

9.2.4 Overview

Making substantial advances on these fronts poses formidable technical challenges, and will surely take a long time. If achieved, however, such advances will automatically bring a much higher level of domestication. In addition, some advances, especially in in-vitro propagation, will enable or facilitate the use of new technologies that allow greater and faster genetic improvement than ever before. That, however, is likely to involve genetic improvement of properties that were hitherto intractable for tree breeding, now that some of the most easily achievable gains have already been secured. Such developments will further raise the level of domestication. Some will save labour and costs. Others, however, may entail additional costs, but domestication is all about being prepared to make greater inputs in order to reap better returns. Pursuit of higher returns is liable to incur increased risks. The risks are probably manageable, technically. Concerns remain, however, over how well institutions will manage such risks (Burdon 2010b; Dungey et al. 2015).

9.3 Main Issues and Drivers of the Future

Many drivers of change exist. Human population pressures and the associated need to increase the production or utilization of renewable resources will be paramount. Within this context, many drivers of current or future change can be listed and outlined, under the following major issues:

• What will the species be grown for?

• Where will it be grown?

• How it will be grown?

• What will be the course of continuing genetic improvement?

• What will be the impacts of institutional and political factors?

Some of the drivers, while listed under particular subheadings, will have implications for one or more others. Having listed the drivers, we will then look at the likely changes, under the same subheadings. Cutting across the technical and purely economic issues are political and institutional issues.

9.3.1 What Will the Species be Grown for?

Likely drivers are:

• flux in future markets: uncertain emphasis on different product classes

• advances in processing and utilisation technology, including new bioproducts

• alternative sources of wood and other biomass, potentially affecting comparative advantage of radiata

• energy demands and bioenergy options

• carbon forestry revenues

• other environmental services.

Much debate has centred on the future importance of solid-wood products (traditionally sawtimber, plywood, poles) relative to pulp and reconstituted-wood products. Many have predicted that solid-wood products will decline in importance. Yet, for softwoods, the markets for solid-wood products are what generally continue to make plantation forestry profitable. This profitability is likely to be enhanced by adopting new categories of solid-wood products such as laminated veneer lumber (LVL) and cross laminated timber (CLT). When compared with certain hardwood species, radiata, because of its wood properties, the lack of super-fast site occupation and thus the relatively slow culmination of productivity, and the size to which logs can be grown quite quickly, looks to have a future comparative advantage for solid-wood products.

For the most part, pulp- and reconstituted-wood products are seen as future by-products of growing radiata for solid-wood products. This is especially so now that short-fibred hardwood pulps are becoming increasingly valued for providing high-quality printing papers. Among solid-wood products, there are ones for structural purposes and others for appearance purposes. While structural products often dominate current production, it can be difficult on some sites to produce high-quality structural timber from radiata. Moreover, appearance products can command the real price premiums.

Use of wood for bioplastic feedstocks has also been mooted. While this may become significant, it seems doubtful if much of the future radiata crops would be grown specifically for this purpose, instead of being a means of using arisings from solid-wood crops.

Growing radiata as a dedicated, renewable energy crop looks similarly unattractive. The species lacks the extremely rapid early growth that enables many eucalypts (and some other hardwoods) to produce high yields of biomass on very short rotations. Arisings from producing more traditional wood products, however, are seen as good sources of biomass for energy, although economic production of liquid fuels from wood cellulose has yet to be achieved. Even now, the lignin that is dissolved in chemical pulping is burnt to produce very significant energy for pulp mills, and much available biomass is not traditionally harvested. However, intensive harvesting of such biomass, which can include twigs and even foliage, would almost always deplete soil organic matter as well as mineral nutrients, with potentially serious consequences for future site productivity.

Values of ecosystem services provided by planted forests have been reviewed in a studies by Monge et al. (2016) and Yao et al. (2017, 2017), largely in a context of radiata plantations as a prospective land use. Considering realistic (but non-traded) values of such services, the current economic disadvantages of forest plantations as a land use largely disappeared. We now consider some specific services.

Radiata has advantages for carbon sequestration, which is widely accepted as an important tool for combatting global warming caused by the current rapid build-up of atmospheric carbon dioxide (Whitehead 2011), even though there will be some countervailing effect of afforestation increasing the absorption of sunlight. Advantages of radiata for carbon sequestration are its attractiveness for growing for solid-wood products on longish rotations, its ability to sustain high wood production in relatively old crops, and its ability to thrive on relatively poor soils. This general attraction, however, is a bonus rather than a self-sufficient justification for growing the species, but it has important implications for how the species is likely to be grown (Sect. 9.3.3). In much of New Zealand, growing radiata on rotations of 60 years or longer (Fig. 9.1) for increased carbon sequestration looks technically feasible (Woollons and Manley 2012).

Fig. 9.1

View of 65-year-old radiata stand at Rotorua, New Zealand during harvesting, showing feasibility of long rotations. However, such rotations incur high proportions of unwanted heartwood, and delay sequestration of carbon as wood in service (Photo John R. Moore)

As with carbon sequestration, the value of radiata for soil stabilisation (Fig. 9.2) and moderating flood-peak streamflows, is an important bonus. The advantages for both carbon sequestration and soil stabilisation may actually be more significant in determining locally where the species is grown than in governing how it is then grown for wood products. Situations will presumably continue to exist where planting radiata can achieve other forms of soil rehabilitation, where topsoil has been lost and even some salination is threatening. There will, however, be cases where radiata can cause unacceptable reductions in water yields from catchments. However, log harvesting operations do entail risks of significant mass flows of sediment and logging debris which are environmentally harmful and, even if very uncommon, incur very bad publicity.

Fig. 9.2

Aerial photo showing negligible soil erosion in a basin planted with radiata compared with abundant slipping in an adjoining basin in pasture on fertile but erodible hill country in the North Island of New Zealand (Courtesy PF Olsen Ltd (Back cover New Zealand Journal of Forestry 59(2) 2014))

Large areas of land remain in New Zealand where planting radiata would achieve soil stabilization, although government policy has not been addressing this issue effectively. In Chile, large areas of land remain where afforestation with radiata could be used to rehabilitate degraded soils, which is now being more specifically encouraged by government. In Australia, where salination of farmland is a widespread problem, remediation by planting radiata is unproven and, even if it works, would need to be weighed up against reduced water yields.

Radiata will doubtless retain a role for farm shelter in New Zealand, and to a lesser extent in Australia. Shelterbelts are typically grown to provide substantial volumes of timber as a bonus, effectively occupying for wood production areas well beyond the span of the tree crowns. Sacrifices in potential timber production, however, will be needed in order to ensure continuity of shelter over time.

9.3.2 Where Will the Species Be Grown?

Several main influences are seen:

• pressure on availability of land: demands for agriculture, horticulture and conservation estates

• climate change: warming, storm intensities, increases or decreases in rainfall

• biotic developments: arrival of previously absent pathogens and pests

• availability of other tree species that are commercially competitive with radiata.

Of these, the second and third can be strongly interrelated because of the influence the relationship between hosts and their pathogens and pests, so they are reviewed together. Finally, the expected changes in global and regional distribution of radiata plantations are reviewed.

9.3.2.1 Classes of Land—Availability and Suitability

By and large, the better the land and the easier the terrain, the better forest plantations will perform, although there are exceptions. More importantly, however, plantation forestry must compete for land, economically and politically, with alternative land uses.

Very high quality land may have no special advantage for plantation forestry; indeed, it can have significant disadvantages, apart from being too expensive. Many tree species, especially pines, need only modest soil fertility to grow well, higher fertility being conducive to problems with both tree form and basic wood quality. Many very fertile horticultural soils may also lack the good drainage that radiata needs. For other crops, both agricultural and horticultural, the advantages of high soil fertility and/or easy terrain are often much more crucial than for growing forest plantations. And there is land where climates are well suited to horticulture, such as growing grapes and other fruits, but unsuited to intensive wood production. In the long term, land of generally modest fertility and often of rolling to steep terrain will remain available for growing radiata, despite the greater difficulties and higher costs of harvesting. There may be areas where forest cover reduces catchment water yields, through increased interception and transpiration losses, so much as to lead to prohibition of growing radiata (cf Mead, p. 48).

In New Zealand, complex land-use issues are involved. There some recent changes in land use have occurred, with thousands of hectares of radiata forest being converted into dairy farms. This happened on easy terrain, with a boom in dairy prices, and on soils where such a conversion was relatively easy. Also, the regulatory machinery was ill-equipped to review and restrict such land-use changes in order to head off likely environmental ill-effects. On the other hand, large areas of marginally economic pastoral farmland may be converted to radiata, especially with incentives for carbon sequestration (Levack 2016). The demands on land for commercial production in some areas, and for conservation values in others, will tend to favour intensification of growing practices for commercial crops, although the type of land that will be available for forest plantations will tend to militate against that. The net effects of this trade-off remain to be seen, but are likely to vary from case to case. A complicating factor in New Zealand is the price regime for marginal land for sheep and beef farming. Despite the marginal status, the land prices are effectively governed by those for pastoral farming, yet those prices bear little relationship to current profitability, settling to values that mean returns of 2% or less on the farming operation, well below normal commercial discount rates. The prices that get paid presumably reflect a combination of two interdependent factors: preparedness to pay a premium for the right to run one’s own farming operation and an expectation that this will ensure capital gains upon selling the properties. Since such land prices greatly affect forest profitability (Evison 2008, 2013; Richards 2013), afforestation of such land may need some active incentives in the absence of a functional carbon-credit market. Recent volatility of dairy commodity markets may help with market corrections, but not in the longer time frames needed for orderly commercial forestry. Moreover, any government failure to impose “polluter pays” will militate against both timely carbon sequestration to mitigate climate change and maintaining water quality in streams and lakes.

Political antipathy can exist within rural communities to both overseas interests and local urban-based investors owning large land holdings. Maori ownership of large areas of land currently or potentially carrying plantation forest might be expected to favour plantation forestry. Yet some of the owners want such land to revert to native forest, essentially for its cultural value. Probably more significant, though, is an upsurge of interest in manuka, a native scrub species, for producing honey or essential oil (Orme 2017). Long prized for its flavor, that honey is now even more prized for pharmaceutical properties, leading to a boom in research on its genetics and culture, and to various speculative ventures. Should such ventures succeed well, much land could be diverted from commercial forestry. Otherwise, much time could be lost in rationalising land use.

9.3.2.2 Climate Change and Biotic Developments

Where radiata can now be grown is well understood, because its attributes have led to it being tried so widely. Mild temperate climates, away from tropical and subtropical summer-rainfall conditions, are seen as where radiata will be grown in the future. Radical changes of its global distribution, even with foreseen climate change, seem unlikely. Some changes, however, may become possible if radiata’s biotic tolerances can be further improved by being able to hybridise it with additional species, or being able to use genetic engineering, thereby extending or at least safeguarding its current range.

In many areas where radiata is grown on a large scale, some known and often serious fungal pathogens are not yet present. Western gall rust (WGR, caused by Endocronartium harknessii J.P.Moore) is a case in point. The likelihood of it being introduced and becoming established in Southern Hemisphere countries is greatly reduced by the strong asynchrony of infection seasons between hemispheres. If, however, WGR did cross the Equator there would almost certainly be large areas where it would be extremely dangerous, being likely to cause much deformation of the lower boles or even tree mortality. WGR can be devastating on radiata north of radiata’s natural range in California, where conditions are moister, and within the natural stands of radiata WGR is especially prevalent after unusually wet springs, which roughly match humidity regimes that prevail over large areas where radiata does well as an exotic. Since WGR is associated with the juvenile or semi-juvenile growth phase, its impact should be mitigated by growing material of provenance origin that shows more rapid onset of maturation (cf Old et al. 1986) or vegetative propagules (cuttings or plantlets) with appreciable maturation (cf Zagory and Libby 1995).

Globalisation of trade, and even of tourism, will continue to increase the risk of new pathogens or pests becoming established in grower countries, despite the best efforts in border biosecurity. Yet despite the known existence of hazardous pathogens and pests, the biotic alarms with radiata have mostly arisen with pathogens and even insect pests with no prior history of being serious.

Overall, climate change is seen more as accentuating certain biotic hazards, and mitigating some others, rather than of itself creating completely new ones.

9.3.2.3 Competing Species for Plantation-Forestry Roles

Even where radiata will continue to thrive, other species will be competing for parts of its niche as a plantation-forestry species. A number of eucalypt species have been gaining increasing popularity for plantation forestry, even displacing radiata in parts as the species of choice. Factors include: the very rapid early growth of various eucalypts, the associated amenability to growing as short-rotation pulpwood crops, the economic attraction of very short rotations, the favoured status that their short-fibred pulp has acquired, and the continuing “biotic honeymoons” of freedom from certain insect pests enjoyed by eucalypts on land masses remote from Australia. Countervailing factors for eucalypts include the site sensitivity often shown by them; the associated preference for growing them on easy country, often land with a horticultural potential, compared with growing them on steep sites with often low soil fertility; the limited long-term carbon sequestration under short-rotation pulpwood regimes; the difficulties for satisfactory conversion to solid-wood products that are posed largely by growth stresses; and the prospect of biotic honeymoons ending. Indeed, the biotic honeymoon for some eucalypts in New Zealand ended around a century ago with insect pests becoming established from Australia across the Tasman Sea (“The Ditch”), a process that continues now (Kay 2005) and doubtless into the future. Establishment of insect pests, however, can be slower than that of the fungal pathogens that have often affected radiata. Moreover, research is proceeding to improve eucalypts genetically for producing solid-wood products. Problematic, however, is the future impact on eucalypts of myrtle rust which is caused by a South American fungus that is spreading rapidly in some other parts of the world.

More locally, coast redwood could gain increasing favour, at least in New Zealand, as an alternative to radiata on some site categories. Its long-term productivity can be very high, it has very few pests and diseases, its timber can be of high value, and its coppicing behaviour can both avert replanting costs and ensure continued soil stabilisation after harvest. However, it tends to be much more site-demanding, is more difficult and expensive to establish initially, and requires longer rotations to produce quality timber. In New Zealand, Douglas-fir has a major advantage for some areas in much superior resistance to snow damage.

9.3.2.4 Prospective Shifts in Distribution

Some significant shifts in the distribution of radiata plantations are expected both within and among grower countries. Climate change could bite at both levels. Within Australia, likely shifts have been addressed by Booth (1990), Booth et al. (2002) and Pinkard et al. (2010), and within New Zealand by Kirschbaum et al. (2012) and Watt et al. (2012), through either direct climatic effects or effects of pathogens and pests.

The likely climate-related changes are various, with both local contractions and local extensions of plantation areas. Some areas will become too dry, and better for, say, olive growing. Other areas, however, may become too warm and humid, and/or subject to violent, damaging storms. On the plus side, there are expected to be areas where rainfall increases to acceptable levels without undue heat and humidity, or where the present-day risks of frost or snow damage become acceptably low.

Among countries, changes could be additionally driven by other factors, notably: uncontained incursions of pests and diseases, whether pathogens have effective vectors; and political and institutional factors.

9.3.3 How Will the Species Be Grown?

A number of likely drivers are involved here:

• interplay between genetic improvement and nursery production systems

• environmental issues: sustainability of site productivity, biodiversity, public reaction

• advances in plant protection: new-generation herbicides, new biocontrol options

• biotic developments: unpredictable spread of diseases and pests, but new control options

• precision field technology

• carbon forestry: own economics, impacts on stocking and rotation ages

• cost structures: manual operations more expensive, favouring mechanization

• advances in crop modeling techniques.

Plantation forestry, growing even-aged crops that are clearfelled, is expected to remain the norm for radiata. While clearfelling draws much public criticism, routine implementation of uneven-aged, continuous-canopy silviculture is generally highly impractical with a species like radiata. Indeed, clearfelling and replacement by even-aged stands mimics the natural regeneration ecology better. The norm in nature is evidently for fires to kill significant areas of radiata forest which get replaced after fire causes release of the cumulative seed bank in the closed cones. Replacement stands are typically much healthier than ones resulting from regeneration under an existing canopy. In the bigger picture, successful domestication of forest trees has been predominantly with species that are at least in part of pioneer status ecologically.

Artificial regeneration is generally a central feature of modern plantation forestry. In radiata, use of bare-rooted nursery stock has been greatly refined, yet there have been recent moves towards use of container-grown stock to help achieve quicker and more efficient capture of genetic gain from clonal selection, with a bonus of extending the safe planting season. This is now being implemented in some major nurseries in Chile and New Zealand, at least. Its adoption creates a strong incentive to develop mechanization in what is largely a new area.

Clearfelling can certainly create problems with soil erosion after harvest, but these are typically transient, and so affect only very small portions of a forest estate at any one time. Likewise, elevated but transient leakage of soil nutrients following clearfelling typically affects only small sub-catchments at any given time. Surprisingly, clearcut areas can perform an important role in maintaining landscape biodiversity. They produce favoured habitat for certain bird species, a specific case involving radiata being the endangered but iconic New Zealand falcon (Seaton et al. 2009). Moreover, radiata plantations even support populations of the rare New Zealand long-tailed bat (Borkin and Parsons 2010), and some very dense populations of the kiwi (Kleinpaste 1990), although averting ill-effects of harvesting operations poses challenges.

Control of competing vegetation, namely weeds, is likely to become more of an issue in the future, involving plantation forests in general and not just radiata. Typically, weeds are easily dispersed and far more easily established than eliminated, creating the prospect of an ever-increasing suite of weed species that depress tree growth by competition and/or inflate plantation-growing costs. Mechanised clearing of vegetation will remain an option, but it must be used carefully on steep terrain, and even on easy terrain it can entail scalping of topsoil. A more strategic approach, timely control of incipient weed infestations, often brings costs and criticism from anti-herbicide activists but no immediate profits for forest owners nor plaudits for the managers. More weed species will tend to create a need for broader-spectrum herbicides, and yet various traditional herbicides have been outlawed or likely to become so. Genetic engineering, to incorporate resistance to broad-spectrum, environmentally-benign herbicides remains a possibility, and its feasibility has already been demonstrated (Bishop-Hurley et al. 2001). Although evolution of herbicide tolerance in weeds is emerging as a problem with agricultural cropping, this seems much less of a danger in forest plantations where the need for weed control is typically for short periods at the start of each rotation. Future use of higher stockings may be justified in part by how it can help control weeds for current access and to ease crop re-establishment.

Use of fire to remove logging residues and help control weed growth has been widely discontinued, while fire can actually favour some troublesome weeds. Such burning provokes public objections, to both the smoke and perceptions of ecological ill-effects. Apart from the smoke pollution, burning entails immediate losses of nutrients and release of carbon dioxide. Yet fire, despite the immediate loss of site nutrients that it entails, may be conducive to sustained soil health and fertility, through generating soil charcoal or biochar. Indeed, fire is part of the natural ecology of pines in general, and the trade-offs in its exclusion or use are unclear.

Control of pests and diseases will surely remain a very important issue, given the exotic status of radiata plantations, how modern transport facilitates incursions, and the attractions of growing radiata where it will perform at all acceptably. Genetic resistance would be preferred, if available within a reasonable time and if proof against genetic shifts in pests or pathogens. Chemical control, as with spraying, may well retain a role. Biological control, however, has big technical and ethical attractions. Insect pests are often amenable to biological control, and this has already been implemented in radiata (e.g. for sirex—Kaya et al. 1993; Burdon 2001). In addition, there are exciting prospects for biological control of fungal pathogens. In horticulture, applying spore suspensions of certain fungi can already control others. It is now known that conifer tissues can contain rich suites of endophyte fungi, of which at least some may confer resistance to pests and diseases. Manipulating these populations may afford future biological control (Ganley 2010; Burdon 2011).

There will almost certainly remain some conflicts arising from the fact that on the more productive sites, with relatively high rainfall, risks of fungal diseases will tend to be elevated (Burdon 2010b), in a classic risk-reward relationship. But on drier sites, where disease is less of a hazard but inherent productivity lower, insect pests will tend to be more of a hazard.

Use of fertilizers is likely to be more significant in future, especially to replace nutrients that will be lost in the course of more complete harvesting and use of biomass. Precision systems should make possible better-targeted applications. Raising general fertility, with elevated nitrogen availability, can enhance wood volume production much more than height growth. This accords with classical experimental findings that optimum nitrogen levels for diameter growth exceed those for height growth in pines (e.g. Roberds et al. 1976). Actually, height growth and stem volume production can be significantly uncoupled among regions in radiata, with lower optimum temperatures for volume production (Burdon 2001; Palmer et al. 2010), and in loblolly pine (DeBell et al. 1989). Also, using fertilizers to improve growth without unduly compromising wood properties appears to be a challenge on at least some sites. Indeed, the availability and desirability of using fertiliser for growing radiata can pose some major conundrums. As a true pine, radiata can tolerate relatively low soil fertility, which will commend it to relatively low-quality land. Nevertheless, productivity of radiata can surely be boosted substantially over large areas by raising soil fertility, yet requiring much less fertility than most food and pasture crops, meaning that water quality in streams and lakes would not necessarily be jeopardised by using fertiliser in plantations. Indeed, the finite area of available land will argue strongly for raising soil fertility for growing radiata if it is at all economic. On the other hand, pines appear to be generally ill-adapted to high soil fertility in respect of wood quality, so the fertility levels needed for optimal productivity of radiata are very inimical to wood quality. Thus, for substantial areas of land a major challenge may lie in genetic manipulation of wood properties such that wood quality remains good even under elevated fertility.

The issues of wood quality pre-eminently involve the corewood zone, which in radiata will pose ongoing challenges (Moore and Cown 2017), not only for the forest managers abut also for the tree breeders.

The emerging (but sputtering) market for carbon sequestration favours both the adoption of higher stockings and backing off from very short rotations. This, however, is largely reinforcing trends that have been emerging from the realization that low stockings and shorter rotations have been compromising both yields and wood properties to a greater degree than had been assumed. Another reinforcing factor is the increased role of USA-based TIMOs (timber investment management organisations) for whom economically optimal rotations are longer than for conventional corporates.

Other aspects of management regimes will depend strongly on anticipated markets and end-uses. Very problematic is the future of pruning in order to produce clear timber. As a highly manual operation it is becoming relatively more and more expensive. Many forest growers, who see the costs, but not the final returns, are flinching from this operation, especially in the light of recent market behaviour (a weak market in USA for high-quality clearwood, but a strong market in China for knotty logs). Moreover, the sacrifice in wood volume with pruning accentuates its economic dependence on strong premiums for clear timber. Modifying the branching pattern by genetic selection is an option, but has little promise for the bottom 3 m of the stem, where branching behaviour tends to rule out high-quality clearwood in unpruned stems but wood properties militate against quality structural timber (Burdon et al. 2004). Pruning to just 3 m would fit well with the proposal (e.g. Xu and Walker 2004) to segregate as a separate log the bottom 3 m of the bole, but this would pose problems for harvesting, logmaking and log transport. In Chile, the continuing level of attack by the pine shoot moth (Mead 2013) will now militate against growing trees of the “long-internode” branching habit which recover less readily from leader damage.

Underpinning many of the future field operations will be adaptations of “precision agriculture” technology. Obvious examples—and only examples—will include navigational aids for aircraft that are used for spraying and applying fertilizers. Use of remote-sensing technologies is being much aided by the availability of small Unmanned Aerial Vehicles (UAVs or “drones”); this allows detailed ground profiling of forests which is extremely valuable for forest management and potentially very useful for tree breeding.

Growing clonal material has important attractions, provided risk-spread multi-clonal portfolios are deployed and grown judiciously. Clonal forestry, as it is called, offers greater potential genetic gains. It can also be used to avoid unwanted genetic variation. Avoiding such variation can substantially simplify processing and marketing. In addition, appropriate use of genetic uniformity should allow far better control of branching pattern, with benefits for clear timber production (Burdon 2008). More radically, it may serve the production of crops that are even better producers of wood (Sect. 9.2.3), especially if clones are deployed as mosaics of monoclonal blocks.

Straddling the issues of how the species will be grown and what genetic improvement can be sought and achieved is modelling of crop performance, for predicting outturns under alternative growing regimes and processing options. Such modelling has advanced dramatically in the last few years, with increased computing power and advances in remote-sensing technologies (Sect. 7.3.3). It will doubtless be subject to further refinement, leading to better-informed decisions by forest managers. Challenges will remain in adapting the models to accommodate continuing changes in genetic populations. More fundamentally, better models, based largely on better genetic information, should serve to project genetic gain to harvest age and quantify, at the level of the crop, the trade-offs involving adversely correlated traits. Indeed, quantifying such trade-offs satisfactorily can often be crucial for defining appropriate breeding goals as well as selecting the appropriate individuals.

9.3.4 Genetic Improvement Issues

Likely drivers are as follows:

• evolving breeding goals

• advances in assay technologies, for example wood properties and disease or pest resistance

• advances in data analysis, largely incremental

• advances in gene technology: pedigree reconstruction, selection tools, hybridization techniques, genetic engineering

• the bioinformatic revolution

• advances in propagation technology: mass-propagation systems, in-vitro technology, control of maturation state, hybridization, platform for genetic engineering, control of flowering.

9.3.4.1 Evolution of Breeding Objectives

Future breeding objectives must include pursuit of growth rate, good tree form, desirable wood properties and resistance to known diseases or pests that affect productivity and/or raw-material quality. Beyond that, unless the trees are grown on extremely short rotations, the tree breeders and silviculturalists are facing major uncertainties in both the biology and the markets.

Biological uncertainties are various. They include:

• imperfect knowledge of the genetic variation in traits of potential interest

• often very imperfect knowledge of genetic correlations between different traits, which can be particularly important with adverse genetic correlations

• limits to what are realistic breeding goals, including those imposed by adverse genetic correlations

• detailed patterns and drivers of genotype-environment interaction

• biotic events, such as the arrival of new pathogens, which may impose new breeding goals, noting that major biotic events affecting radiata have repeatedly involved pathogens with no prior record of being significant

• potential side-effects on field fitness (e.g. mechanical stability of standing trees) of certain improvements in wood properties

• impacts of impending climatic changes.

Market uncertainties are largely self-evident. Consumer fashions may change, while the economic fortunes of client countries that have their own consumer preferences may fluctuate. The development of alternative or substitute products may depress demand, or else declining supplies of alternatives can boost demand. New processing technologies can produce market competitors, or they may boost the value of one’s own wood outturn.

Despite the uncertainties, certain breeding objectives are undisputed. Faster growth is always welcome, provided it is not unduly at the expense of wood quality. Straight, healthy trees, free of malformation are always wanted, too. Improvement of wood properties, especially stiffness and dimensional stability, is generally very welcome, most of all in the corewood zone. Fortunately, low microfibril angles are conducive to both good stiffness and dimensional stability, which are especially desirable for structural and appearance products alike. That said, carbon forestry, by lengthening optimal rotations, will improve average wood properties substantially, an exception being increased heartwood content which is not wanted by processors of radiata wood (Kennedy et al. 2013).

More debatable is the optimum branching pattern (Burdon 2008). Future price premiums for radiata clearwood are uncertain despite some good current premiums for appearance-grade timber, pruning costs are becoming problematic, and yet production of clearwood without pruning faces biological constraints. While the short-internode branching pattern confers superior adaptation to many New Zealand and Chilean sites, it is not conducive to obtaining good yields of clear timber as clearcuttings without pruning. Yet the other extreme of branching pattern, namely a very long-internode habit, tends to incur penalties in growth rate and tree form (Burdon 1992). If there is an “intermediate optimum” for branching habit, which gives good yields of clearcuttings without much sacrifice of potential growth rate and tree form, it should be achievable through clonal forestry.

Gains in productivity, while always welcome in themselves, are even more so in a context of carbon forestry.

Between the biological uncertainties, market uncertainties, often complex production systems for processing and their ramifications, choice of appropriate strategies and algorithms for genetic selection poses very complex challenges (Ivković et al. 2010a, b), especially for radiata. On these challenges, the last word will not be said anytime soon.

9.3.4.2 Assay Technologies

Improvements sought in wood properties often relate to performance in service. Stiffness can now be assayed on standing trees (Grabianowski et al. 2006), bypassing problems of obtaining cheap, large-scale assays for microfibril angle, and there have been advances in achieving cheap, non-destructive assays for other key wood properties. Similar improvements can be expected in assays for growth performance, climatic tolerances, and disease or pest resistance. Even if DNA sequences eventually become the basis for selecting for these various attributes, this will almost certainly depend on enhanced assays of the actual trees as intermediate steps in developing applications of the gene technology. Reliable cross-reference between field performance and individual genetic variants (represented in DNA polymorphisms) will be essential for such selection, but a more immediate task will be developing assays that are good proxies for field performance.

9.3.4.3 Classical Data Analysis

Data analysis for field genetic experiments, and other empirical trials, has undergone major advances in recent years. Even so, further, incremental advances can be expected, especially as computing facilities become still more powerful. Integrating data on physical performance of trees with DNA data will be the target of much future effort.

9.3.4.4 Use of Gene Technologies

Gene technologies have several potential roles in genetic improvement. These include:

• 1. Verifying or discovering genetic identity

• 2. Providing a basis for selection, either in lieu of or in conjunction with assessing field performance and all the delays this can entail

• 3. Helping identify feasible breeding goals

• 4. New hybridization technology

• 5. Use of genetic engineering (GE), either operationally or to inform conventional breeding.

Of these, (2) and (5) represent particularly fast-evolving fields of research and technology, making any pronouncements especially tentative.

Verifying genetic identity is already a well-developed and widely-proven technology (Burdon et al. 2008). It can avoid much loss of genetic gain, by avoiding use of wrong clones in seed orchards or of crosses coming from the wrong parents. For the future, there is the prospect of identifying the pollen parents of individuals resulting from natural pollination, thereby maintaining the advantages of full pedigree without the labour and costs of traditional controlled crossing. That allows better assessment of the genetic merit of selection candidates for the next generation, control over unwanted inbreeding, and much better maintenance of genetic diversity for the future. Actually, there is a potential to improve upon classical pedigree information through being able to pick up random departures from the expectation that an individual receives one-quarter of its genes from each grandparent.

Being able to select individuals on the basis of genes has the attractions of saving on costs of expensive field trials and saving time spent waiting for reliable field expression of traits. That attraction is become greater with steadily decreasing costs of obtaining DNA sequence information. However, this depends on relationships being established between DNA information and field performance, which itself can take a long time and involve very large numbers of individuals which may be very costly to produce (Burdon and Wilcox 2011). Hopefully, this process can be shortened by narrowing down the list of possible genes of interest, using candidate genes that have been shown, very often in other types of plant, to exert effects of interest.

However, recent experience with animal breeding, and the dramatic reduction in costs of DNA sequencing, have shifted the emphasis towards genomic selection (Hayes et al. 2009a; Grattapaglia and Resende 2011; De Mello et al. 2014). Rather than searching for and trying to verify quantitative trait loci of large effects, the emphasis is now on genome-wide screening to use for selection the sequences in chromosomal segments associated with phenotypic effects. Also of value is the marker information provided on the genomic contributions from various ancestors (Realised Relationship Matrices, e.g. Hayes et al. 2009b). Already used for large-scale dairy-herd breeding (Sect. 9.2.1), genomic selection has potentially worthwhile reliability for a species like radiata as soon as embryos or seedlings can be genomically assayed non-destructively. Even if the tree breeder remains dependent on some field evaluation of selection candidates, genomic selection may help greatly in shortlisting the candidates. Taking full advantage of such a capability, however, would only become feasible with greatly accelerated sexual reproduction, on a basis of “flowering on command and command only.”

Major challenges remain, though, for the tree breeder pursuing genomic selection (cf Nakaya and Isobe 2016). For large, modern dairy-breeding programmes, commercial herds are the breeding populations, and abundant good phenotypic data on individuals come available from the monitoring now entailed in routine herd management. For tree breeding, in which breeding populations are separate from commercial crops, comparable data are certainly not obtainable in the course of routine forest management, and for some traits good phenotypic data can take considerable time to become available. Improved remote-sensing technologies are likely to help, though. Limited linkage disequilibrium in largely undomesticated trees, together with the large conifer genomes (Sect. 9.2.1), pose additional complications, which may restrict satisfactory application of genomic selection to only small subsets of forest-tree breeding populations.

What breeding goals are feasible will depend very much on pathways of gene action, which can now be traced in much greater detail. Of special importance are cases involving traits that are adversely correlated. If the adverse genetic correlations between traits involve fundamentally antagonistic effects of genes exerting desired effects on the different traits, the scope for simultaneous gain in the traits may be severely restricted, meaning that solutions other than genetic improvement may need to be sought. A specific question, affecting radiata along with other conifers, is whether producing the syringyl rather than guaiacyl lignin, which makes for the easy pulping of broadleaved trees, is compatible with the rest of conifer metabolism. This objective, however, is not accessible through conventional breeding, but rather through genetic engineering (Sect. 9.3.4.6).

New hybridization technology may make available a new range of genetic diversity for future genetic improvement of radiata. At present, it is crossable with very few other species. Among those species is knobcone pine, Pinus attenuata. Hybrids with knobcone pine, while very susceptible to dothistroma needle blight, are far more resistant than radiata to snow damage (Dungey et al. 2011). Such resistance promises a major extension of where radiata can be safely grown. Among the suite of Mexican pines, a few can be crossed with radiata with some difficulty. Several others, which are also closely related, could confer valuable attributes if hybridization could be achieved. In particular, those Mexican pines, coming from summer-rainfall areas, will have very different spectra of disease resistance compared with radiata, which could be crucial in the event of a biotic crisis affecting radiata. However, recourse to using such species in areas where radiata is now grown will require much more research commitment than is generally being made at present.

Genetic engineering (GE) has several attractions. It can be used to confer attributes that are not available from the existing variation within a species. It also brings in very short and highly specific sequences of DNA, instead of the unknown and largely unwanted genetic “baggage” that comes when hybridization is practiced in order to bring in some new attribute. Indeed, it could be an alternative to actual hybridization as a means of incorporating attributes from other species. However, achieving the appropriate specificity depends on knowing the exact sequences that are needed. The basic feasibility of GE with radiata has been well demonstrated (Sect. 7.11.2). As an operational tool, however, it is still unproven, and the lack of unintended adverse effects on the trees and their products remains to be confirmed (Ahuja 2014), although such risks would appear very low for herbicide resistance. Regulatory restrictions on field release of genetically engineered forest-tree material remain a huge constraint (Strauss et al. 2015), adding enormously to development costs. These restrictions are partly driven by political resistance that will not disappear at all readily. Less contentious, perhaps, is the potential of GE as a research tool to inform more conventional breeding. The scope of using GE, and the issues associated with is use, are addressed in more detail in Sect. 9.3.4.6.

Both selection on the basis of DNA sequences and targeting of GE will likely be helped greatly by much research on gene expression, whereby relationships can be established between DNA sequences and properties of the actual trees. This, however, could be hampered by the running down of a substantial workforce engaged on the physiology of radiata during the 1980s.

Use of gene technologies offers the prospect of avoiding many laborious activities in genetic improvement, and greatly foreshortening certain time frames. Yet developing such technologies will, at least in the short to medium term, entail a big increase in commitment to genetic improvement, unless it is done much at the expense of classical breeding activity. Using such gene technologies further intensifies domestication, and domestication is all about increasing inputs in order to achieve better returns.

With advances in gene technology, and with accumulated experience, the scope for GE to achieve environmental as well as economic benefits is becoming increasingly clear (Sect. 9.3.4.6).

9.3.4.5 DNA Sequencing and Bioinformatic Advances

The costs of DNA sequencing are falling dramatically, which opens up huge possibilities for using DNA technology in genetic improvement, in identifying feasible breeding goals, genetic selection, and targeting genetic engineering. Optimism, however, must be tempered by the challenges of genomic sequencing in a species like radiata and of managing and analyzing the prodigious amounts of DNA sequence data that can be increasingly generated.

Radiata, like other pines and conifers in general, has a huge genome (Sect. 7.11.1). Why this should be so is a mystery. Anyway, attempting a complete genomic sequencing is currently prohibitive, and interpreting a complete sequence is a massive challenge. There is, however, something in our favour (Ritland et al. 2011). The genomes of pines, indeed Pinaceae in general, are remarkably similar. The chromosomes are much the same, and the various genes do essentially the same things, and are arranged in essentially the same chromosomes and in very similar sites within the chromosomes. So genomic information obtained in one pine species can be very helpful for studying genomes of other pines. Indeed, progress has been made in cross-referencing genomic information between radiata, loblolly pine (Pinus taeda) from south-eastern USA, and maritime pine (P. pinaster) in France, the work on each species drawing in its own research funding. While loblolly pine can attract the biggest funding, radiata may otherwise have greater overall advantages as a “model species” for research. In fact, there is no need to choose a single pine as a model species for research, because several can be used in conjunction to exploit all their individual advantages.

Dealing with genomic sequence data is termed bioinformatics. Managing and analyzing the sequence data that will come, and interpreting the findings, are already becoming far bigger tasks than just acquiring the data. The increasing availability and dwindling costs of computing power will help greatly, but devising efficient systems for managing and processing the data will still be a great challenge.

9.3.4.6 Role of Genetic Engineering

Operational uses of GE in radiata will presumably come in time, but the nature of the uses seems uncertain. A classic use, conferring herbicide resistance, is problematic, because radiata is comparatively tolerant of weed competition at establishment, and it also tolerates some herbicides which can therefore operate selectively. Conferring resistance to insect pests is a popular target for GE in some plants, notably cotton. In radiata it has been sought as a solution to a problem in Chile, but without achieving practical application; moreover, insect pests are often amenable to biological control. Conferring reproductive sterility, avoiding all diversion of resources into reproduction, is very attractive in principle (Burdon and Libby 2006), and promises the huge benefit of containing the transgenes, which addresses an important regulatory barrier to use of GE. Whether sterility can be conferred completely and reliably is contentious. Also uncertain would be the ecological impacts of preventing the dispersal of mineral nutrients that occurs through the wind dispersal of pollen grain. The ability to induce super-precocious flowering (cf Kean 2010) in material that is otherwise sterile would be extremely welcome for the breeder, but achieving and controlling it presents an additional challenge.

Another research target for GE in radiata is producing syringyl lignin in place of guaiacil lignin, which would make chemical pulping easier and cheaper. While it has been achieved in the laboratory (Wagner et al. 2015) much remains to prove that it can be done without appreciable ill-effects on field fitness and the suitability for solid-wood products.

Incorporating resistance to certain fungal diseases may be yet another target for use of GE. Ideally, one wants to incorporate resistance genes of large effect. One also wants a diversity of resistance mechanisms, “pyramiding” resistance genes (Burdon and Wilcox 2011), in order to ensure durability of resistance against emergence of new and more dangerous strains of the pathogen(s). Achieving this, however, will depend on big advances in knowledge of resistance mechanisms.

Operational use of GE is foreseen in at least the medium term in the context of clonal forestry (Burdon and Lstibŭrek 2010), using for commercial plantations finite numbers of well-characterised clones containing the desired transgenes. This contrasts with the situation with annual crops like soybean, cotton and canola, in which the transgenes are stably integrated such that the crops “breed true” for the genetic transformations. In forest trees, reaching the point of breeding true can take a long time, and the need for it can be obviated by clonal propagation. However, advances in DNA technology now promise to remove this constraint.

Genetic engineering currently faces vehement political opposition from “green” organisations, largely on environmental grounds in the case of forest trees. We are concerned, though, that such opposition is misplaced, to the point of being downright counterproductive to central causes of those groups. Some key goals of GE are surely environmentally beneficial, in direct and indirect ways. Among the more direct benefits are the use of GE-conferred insect resistance to avert the need for objectionable insecticides or of herbicide resistance to favour the use of low-toxicity, non-persistent herbicides. Indirect environmental benefits should accrue from greater and more reliable production on the land area devoted to plantation forestry. While much has been made of the postulated risks of GE, those risks have generally not materialized in closely monitored situations (e.g. Häggman et al. 2014). Moreover, recent advances in gene technology have made it possible to achieve GE on a much more targeted basis, and with interventions that now avoid most of the perceived risk factors. Indeed, it is now becoming possible to release GE crop or fruit cultivars in which extraneous DNA sequences have been excised (Kean 2010). Still more recently, “gene editing“ with CRISPR technology can be used to completely bypass any use of extraneous DNA sequences (e.g. Dance 2015; Hall 2016). There will remain some risks to be managed with GE, but they must be balanced against the potential for GE to serve as a risk-management tool. While the precautionary principle has been applied to place the onus of proof on safety of GE, proper application of the principle is becoming less one-sided, and is likely to eventually reverse. Thus the use or at least the availability of GE is potentially part of an environmental imperative.

9.3.4.7 Delivery Systems for Genetic Gain

Clonal seed orchards, as originally conceived, are largely obsolete for radiata. Given the ease with which mass vegetative multiplication can be practised in the species, it can be used to deliver the leading-edge genetic gain that can be achieved in small-scale production of seed in crossing archives. New forms of in-vitro propagation not only extend the range of options for mass multiplication of material without repeated recourse to seed production. For instance, when combined with cryopreservation, vegetative propagation can give greater control over maturation which is effectively halted at cryopreservation (liquid-nitrogen) temperatures. In this connection, a favoured form of in-vitro propagation is embryogenesis—dissecting out seed embryos and inducing the resulting culture to produce large numbers of fresh embryos which may be reared through into plantlets. Other in-vitro technologies can be used as alternatives to embryogenesis or in conjunction with it. Pines, however, are not yet as amenable to embryogenesis as spruces are, but that may change.

Mass-propagation of proven clones, namely clonal forestry, can capture greater genetic gains than are possible with propagation by seed, and avoids unwanted genetic variation that will always exist among seedlings. Fully successful clonal forestry ideally requires total control of maturation, which ideally includes being able to rejuvenate material from old trees to the state of a freshly germinated seedling. While that occurs every time a seed is produced, the challenge is to achieve it without the genetic recombination that occurs in the course of seed production. In practice, one would want to rejuvenate propagable material of any tree at any age, which remains to be achieved with radiata along with all other pines. Given the special significance that clonal forestry is likely to assume, we will soon revisit it to address some broader issues that it involves.

In-vitro propagation systems can serve another purpose, in providing material on which GE can be performed. This fits nicely with the fact that operational use of GE with forest trees currently seems best implemented in a context of clonal forestry (Burdon and Lstibŭrek 2010).

9.3.4.8 Optimising Field Deployment

Given the market uncertainties, the large impact of site on important traits such as certain wood properties and tree form and disease incidence, and generally adverse genetic correlations between growth rate and various wood properties, it is unrealistic to deploy the same set of genetic material everywhere. Instead, it is appropriate to deploy genetic portfolios, sets of genetic material that are tailored to different sites and/or different markets. Hopefully, a market-based portfolio may allow the grower to cater efficiently for niche markets that are highly profitable but susceptible to gluts. Any set of deployed material, however, should contain its own risk spread: a minimum of seven unrelated clones has been proposed (Libby 1982), but a figure of 16 or so is widely regarded as significantly safer. There seems no justification, though, for the hundreds of clones that are prescribed by law or regulation in some countries.

9.3.4.9 Strategic Genetic Management

Whatever the potential for use of novel gene technology, it is still foreseen that its use will be superimposed upon a platform of classical breeding programmes. Hence the hierarchy of populations (Sect. 5.5.3), accommodating the trade-off between level of genetic improvement and genetic diversity, will likely remain in some form. To recapitulate, the production populations represent the top of the hierarchy, with the greatest genetic gain but least diversity. They are underpinned by breeding populations, representing somewhat less gain but greater diversity, and which are the “engine room” for achieving continued genetic gain. In turn, breeding populations are underpinned by genetic resources, representing the least gain but the greatest diversity. Maintaining such a hierarchy, as a “metapopulation,” makes it possible to enjoy genetic gains without substantial loss of genetic diversity, thereby protecting very long-term genetic gain, and providing for imposed changes in breeding goals. However, it has its costs, both direct costs and opportunity costs, placing demands on commitment to managing various risks. Even with strong commitment, detailed structures of population hierarchies are likely to change with new technologies. For instance, a high degree of pedigree reconstruction should avert the need for laborious and costly controlled crossing, while some traditional restrictions on moving material between levels in the hierarchy (Sect. 5.5.3) may be lifted.

The very success of genetic improvement is likely to jeopardize conscientious genetic management (Burdon 2010b). The availability of greatly improved stock for commercial planting increases the opportunity costs of maintaining less-improved or largely unimproved material that would be the reservoirs of genetic diversity. Given the degree to which radiata has become domesticated, the genetic stewardship of population management is of special importance. These considerations create a special need for strong political and institutional commitment to forward-looking population management. Yet, as we shall see (Sect. 9.3.5), various political and institutional developments are militating against such commitment.

9.3.5 Institutional and Political Changes and Challenges

The future for growing radiata and its further domestication will be in changing political and institutional contexts, which will be intertwined with changing economic circumstances. It is possible to review such changes and the directions that they are currently taking. While that may give pointers to the future domestication, detailed predictions must be speculative.

Forest ownership and corporate structures, along with research and development (R&D) organizations, are key institutional factors. Changes in them are often partly driven by politics, although development of new technologies can also exert strong influences on the R&D organizations. Non-government organizations (NGOs) can be intensely political.

For safeguarding and improving the productivity, quality and profitability of radiata, the institutional and political picture is a concern. Taking many of the opportunities, and meeting needs mentioned earlier, will require staff of high calibre, with an appropriate mix of skills, operating in well-run organizations that have adequate funding and ongoing commitment. Meeting these requirements will require much lead time, yet some key parts of the skills base are being eroded on a global scale.

9.3.5.1 Forest Ownership

Forest ownership has been undergoing major changes in recent years, and will doubtless continue to do so. In both New Zealand and Australia there has been massive privatization of state-owned forest. This has been driven largely by political doctrine that market forces can solve most problems, and by global pressures to reduce tax takes. Essentially complete in New Zealand, privatisation has some way to go in Australia, where it is likely to proceed further than at present. Since privatization, forest ownership has tended to be volatile, forests often changing hands and sometimes repeatedly. Vertical integration, with forest growing and wood processing vested in the same ownership, has become the exception. Globalisation of business has become a prominent factor, in that forest ownership has often become dominated by large, offshore interests. For such interests, radiata wood can be seen more as a commodity, rather than as a source of high-value products. Moreover, for those interests, managing the risks of depending massively on the one species can often be accommodated by global spread of risk. Even among smaller forest owners, who own a very significant part of the New Zealand forest estate, forestry interests often form parts of risk-spread investment portfolios. The risk is basically one of a biotic crisis, probably some new disease, displacing radiata from much of the range where it now thrives. Such a risk is important, because of potentially dire consequences despite low probability of eventuating. The spread of such risk that is enjoyed by individual forest owners contrasts with the risk exposure of forestry sectors and even the national economy in the case of New Zealand (Burdon 2010b).

In Chile, forest ownership has largely stabilized into the hands of Chilean nationals, while vertical integration is considerable, albeit more within groups of commonly-owned companies rather than within individual members of company groups.

In Spain, the forest ownership is mostly fragmented into small ownerships which, however, enjoy considerable support at various levels of government. Choice of tree species, between radiata and eucalypts, remains volatile.

In South Africa much of the radiata plantation area has gone into conservation estate, where restoration of native vegetation is often being pursued. Between this factor, and problems with pitch canker, this is a country where the future role of radiata may well be sharply reduced.

Economic forces have already influenced the pattern of ownership, and market perceptions will affect how the species is grown. The American market has dominated the demand for high-value appearance-grade products, but in recent years has been weak. Products from lower-grade logs have dominated the booming Chinese market and still dominate a burgeoning Indian market. But it is becoming increasingly unlikely that these Asian markets will continue to be dominated by demand for lower-grade log products. Indeed, growing now for these markets in their present form could prove costly.

9.3.5.2 R&D Institutions

The widespread privatisation of forest estate has been accompanied by institutional changes involving R&D organizations. Such changes have often had much the same political drivers as privatisation of forests, and the subsequent changes in forest ownership have had their own impacts. In addition, new technology has brought its own changes within R&D organizations. Outright and rapid exiting of governments from the R&D has not occurred, but funding has commonly diminished and funding mechanisms have tended to change.

Genetic improvement of radiata has increasingly devolved onto tree breeding associations or companies, such as the Southern Tree Breeding Association in Australia and the Radiata Pine Breeding Company in New Zealand, such that breeding operations are no longer embedded within R&D organisations. Even where government support for genetic improvement and other R&D continues, it may be very much conditional upon industry support. In Australia, such industry support essentially ceased for the now disbanded CSIRO Forest and Forest Products Division. Research arms of state agencies have greatly contracted, and forestry research via Cooperative Research Centres is now discontinued.

Within Chile, R&D capabilities relating the growing, genetic improvement and processing of radiata have become minimally vested in government agencies as such. There is major corporate involvement, in Bioforest within the Arauco group of companies, and within the Mininco group. A high proportion of the actual research is done within various universities, which can pose problems of co-ordination. Uptake of overseas research has been significant. The breeding cooperative is still in existence, but the dominance of its role is now reduced. A return to high level of local ownership of forest resources and processing plant is probably a significant advantage.

In Spain, with the fragmentation of forest ownership, government agencies, especially at the provincial level, but with support from the European Union, are playing prominent roles in R&D. While various avenues of biotechnology are being pursued, the platform of physical breeding-programme infrastructure is limited.

In South Africa breeding work with radiata is now limited, and is likely to remain so, except that efforts are being made to hybridise it with pine species that are resistant to pitch canker. Successful hybridization, or achieving resistance through genetic engineering, could bring a major resurgence of its use.

9.3.5.3 Impacts of New Technology

Developments in DNA technologies, along with budgetary restrictions, have led to a cultural divide between traditional tree breeders and the “molecular” camp. That divide has been exacerbated by a perception among the breeders that molecular work has drained off their funding, with consequent opportunity costs in respect of achieving genetic gain. Managing the problems of communication and associated issues of resource allocation will remain a challenge for some time to come.

The traditional model of cooperative organizations undertaking tree breeding has come under severe stresses. Some of the stresses arise from institutional changes, but advances in technology have played a major part. The availability of new technology has widened the gaps in level of technological engagement among cooperative members, which accentuates temptations for some individual members to appropriate new technology for themselves. Specialised biotechnology companies have arisen, sometimes spun off from forest growers’ own biotechnology arms, and with various fresh alliances formed. Competition among such companies, while typically a spur, can be a barrier to the cooperation on which technological advances can depend heavily. Moreover, proprietary appropriation of biotechnology can contribute to adverse perceptions in a public that is inherently wary of new biotechnologies.

9.3.5.4 Non-Governmental Organisations (NGOs)

NGOs can be very influential politically. They can influence politicians directly, or else indirectly through manipulating measures of public opinion. They can also influence consumer purchases. Various NGOs are heavily involved with environmental causes, and in forestry they have set up schemes of environmental certification of timber, only endorsing timber or wood products that have been assessed as having been produced on an environmentally sustainable basis. But, however laudable their ultimate goals may be, at least some of the NGOs have problems with the phenomenon of plantation forestry. Even acceptance of plantation forestry came reluctantly, and various practices aimed at safeguarding and intensifying wood production are not endorsed. Insofar as such non-endorsement may compromise the role of plantation forestry in relieving pressures of exploitation on natural forest, such stances on the part of NGOs can be highly counterproductive (Sect. 9.3.4.6). Needless to say, the growing of radiata, being in the forefront of plantation forestry, has been heavily embroiled in these issues. Hopefully, the stances of certain NGOs will continue to evolve in directions that serve their goals better.

9.4 The Clonal Forestry Goal

Clonal forestry, namely the commercial deployment of finite numbers of proven clones, has already featured several times in this chapter. However, it warrants special coverage, with some recapitulation, because its implementation is a key part of the highest possible level of domestication. It offers a means of capturing genetic gains that are not otherwise available. These come from favourable non-additive gene effects which are almost all lost in the course of seed production. Some of the most intriguing non-additive gene effects involve epistasis, resulting from interactions between the effects of genes at different loci. Such effects, however, are very hard to quantify and characterise, unless they involve massive effects of particular genes.

The genetic uniformity of clonal material promises major advantages for harvesting and processing. Because wood properties are typically so heritable, genetic uniformity allows big reductions in tree-to-tree variation. Even for growth variables, far better uniformity may be achievable. For stem diameter growth, and thence volume production, the fact that all the non-additive gene effects contribute to heritability with clonal systems means that genetic variation is no longer part of the “noise” variation. In addition, avoiding certain physical interactions that occur between different genotypes can further reduce noise variation. Using clones to minimize variation in stem size, however, will demand excellent practice in the nursery and in field establishment, but the latter may be difficult to achieve on many of the planting sites that will be available. Above all, though, the more predictable the wood properties of individual trees, the more effectively can processing be optimized.

Large-scale clonal forestry with a species like radiata would require risk-spread “portfolios” of clones, as a matter of basic prudence. Deployment of such portfolios will need to be addressed at both local and landscape scales, the latter scale involving greater specialisation of clones for definable site categories. Such matching of clones to sites at once promises greater genetic gains. At the same time, the genetic uniformity does not provide a “cushion” against the presence of some ill-adapted genotypes segregants whose failure can be easily tolerated in natural or artificial thinning of growing stands. This makes especially important good matching of clones to sites, or else identification of clones that are very broadly adapted.

More subtle benefits may come from clonal forestry. Some of the key prospective benefits of clonal forestry involve traits that involve interactions among individual trees. Competitive interactions strongly affect growth of individual stems. Where different clones have different growth curves, much unwanted variation in log size may be difficult or impossible to avoid if different clones are grown in intimate mixture. Moreover, it is in monoclonal blocks that it should be possible to exploit divergences between productivity as a crop and competitive ability. With resistance to certain types of climatic damage, such as wind damage, superiority may not be clearly expressed when clones are grown in mixture, because the fate of individual trees can be so strongly influenced by that of their neighbours.

Empirical evaluation of the performance of individual clones as crops, rather than essentially as competitors, is quite impractical for the potential large numbers of candidate clones. A challenge remains, then, to characterise the “crop ideotype” so that far greater numbers of candidate clones can be evaluated with sufficient reliability on the basis of how they conform to that ideotype.

Several of the theoretical advantages of clonal forestry accrue in monoclonal blocks. Even so, it is not fully clear whether or when mosaics or intimate mixtures of clones will eventually be preferred in radiata. The appropriate choice is likely to depend on specific circumstances, and may change over time.

The technical platforms for fully satisfactory clonal forestry with radiata are not yet totally assured. If any tree, whatever its age, can be rejuvenated at will such that it is immediately amenable to mass vegetative multiplication, we would have a full answer. But such rejuvenation with pines remains to be achieved. Short of that, the capability of using cryopreservation to store individuals in a fully juvenile state and allow indefinite mass-propagation would make clonal forestry much more practicable. Recent progress in these directions (Hargreaves et al. 2009) is very promising.

As mentioned earlier, clonal forestry is seen as the likely context for any commercial implementation of genetic engineering.

In a final note of caution, clonal forestry, by depending on use of proven clones, is inherently behind the leading edge of advancing genetic gain for additive gene effects. Its net benefits must therefore outweigh the significance of this lag. That lag, however, may be avoided by committing every candidate clone to cryopreservation from before or very soon after germination. While this is logistically demanding, storage of individual clones can be discontinued immediately after culling.

9.5 Concluding

The growing of radiata has seen fluctuations in rates of planting and even in total areas of plantation. These fluctuations have resulted from various factors: biotic, political, economic and institutional. In parts of the world, especially southern and eastern Africa, there have been substantial retrenchments of the areas where it is grown, first through diseases catching up with plantations in inherently unsuitable climates, and latterly through introduction of a new disease in the presence of vectors. On the other hand, one major biotic alarm, with Dothistroma in New Zealand, actually strengthened the position of radiata as the preferred species. Also, use of fertilisers reversed some initial plantation failures and substantially extended the range of sites where radiata can thrive. Further biotic alarms will doubtless occur and contribute to fluctuations in planting. Political decisions and wars have contributed more or less directly to booms and lulls in new planting. Less directly, political decisions have led in recent years to changes and volatility in forest ownership, leading to some major lulls in planting. The global financial crisis has also contributed to lulls, partly through a consequent collapse in prices for carbon credits. A boom in short-fibred pulp, along with the financial attractions of growing short-rotation eucalypt crops, and the “biotic honeymoon” of exotic eucalypts, have also eroded or stalled the status of radiata in recent years. Nevertheless, the areas of infertile land remaining, their catchment values, the likely need for solid-wood products, and the advantages of solid-wood crops for carbon sequestration would all point in favour of future expansion of the role of radiata. The outcome of the 2015 UNFCCC conference in Paris, at which the acceptance of climate change as a problem gained international agreement, should lead to a boost in afforestation. Climatic change will surely cause some shifts in the areas where radiata can be grown successfully, rather than necessarily affecting the total area (cf Mead 2013), but genetic improvement should be able to extend the range of suitable climates. The anticipated pluses and minuses for the future role of radiata are summarized in Box 9.1.

Box 9.1 Strengths, Limitations, Threats and Opportunities for Future Role of Radiata

• Basic strengths

  • Outstanding growth potential for conifer, with sustained productivity

  • Significant range of climatic tolerances

  • A pine’s tolerance of limited soil fertility

  • Tolerance of significant weed competition

  • Amenability to selective herbicides

  • Ease of handling in nursery and transplanting

  • General ease of wood processing and utilisation, especially for solid-wood products

  • Genetic variability well sampled and now available

  • Domestication achieved to date

  • Existing scientific knowledge base on species

  • Carbon sequestration potential

• Current limitations

  • Climatic tolerances

  • Biotic vulnerabilities

  • Early relative growth rate still modest

  • Wood-quality problems

  • Diversion of resources into reproduction

  • Competitor, not crop ideotype

  • Imperfect control of maturation

  • Very limited crossability

  • Problems posed by huge genome

• Threat factors and hazards

  • Pest/pathogen incursions: with potential for biotic crises

  • Climatic change: contractions in geographic range; increased risks of climatic damage and biotic crises; extending geographic ranges of pests, pathogens and weeds

  • Potential phosphorus shortage: affecting part of species’ range, but problems subordinate to those of food production

  • Weeds: extended climatic ranges; continuing incursions

  • Biosecurity barriers: restricting access to germplasm

  • Competing land uses: farming (on easy terrain), eucalypt plantations

  • Increased mechanisation: degrading economics of pruning

  • Political pressures: opposition to herbicides and wood preservatives, subsidies to alternative land uses, mis-perceptions of environmental hazards

  • Institutional factors: compromised risk management

  • Genetic improvement: militating against risk management in protecting diversity

  • Economic pressure for earlier harvests: compromised wood properties; advantages accruing to eucalypts; reduced carbon sequestration

• Opportunities: factors and influences

  • Genetic improvement: gains in productivity; improved wood quality; enhanced pest/disease resistance; extended site tolerances (direct and biotic)

  • Propagation technology: allowing accelerated breeding; faster and better capture of genetic gain; more efficient production of planting stock

  • Climatic change: some extensions of geographic range; premium on carbon sequestration, both direct and favouring wood over competing materials; accentuated call for soil and catchment protection

  • Radical upgrade of domestication: flowering on command and command only, redesign of tree architecture; genetic engineering

  • Environmental services: existing calls for soil and catchment protection and rehabilitation; wildlife habitats

  • Enhanced management systems and technology: remote sensing, software systems, nutrient budgeting

  • Precision technology: better application of protective sprays and fertiliser, and other operations, assisted by new remote-sensing systems

  • Novel DNA technology: both informing and supplementing classical breeding

  • Overcoming crossability barriers: radical gains in biotic resistances

  • Biocontrol: of pests, weeds and pathogens

  • Improved biosecurity diagnostics: direct biosecurity gains, better germplasm availability.

Radiata provides a case history of domesticating a forest tree species, in both the achievements so far and the challenges ahead. Both the history and the future embrace a wide range of issues which are varyingly intertwined: scientific and technological; economic, political and institutional; and environmental and even humanitarian. We will not, however, trace these various issues afresh; rather, we just re-emphasise two relating to domestication vis-à-vis the wild state. Domestication of forest trees offends the sensibilities of many who see forests as essentially wild ecosystems. Intensive cultivation is likely to reduce biodiversity where it is practised, although mitigation of losses will often be possible. Of far greater importance is that intensive use of very large areas of land will be an imperative for the world’s human population. The more intensive such use can be, without disastrous environmental consequences, the more land should remain available for recreation and biological conservation. As Brockerhoff et al. (2008) have argued, plantation forest can in various ways play a very positive role for biodiversity. There are direct ways in which the presence of the cultivated tree species can favour organisms whose perpetuation is desired; less directly, minor modifications in plantation management can serve to favour biodiversity and general amenity values; and quite indirectly the availability of plantation-grown wood can relieve at least some exploitation pressures on natural forests. Within radiata, there are the challenges of achieving the benefits of domestication, on the one hand, and maintaining genetic diversity both for its own sake and for providing for the distant future and unpredictable developments, on the other. Reconciling this apparent conflict is quite possible technically. Achieving that, however, faces stern political and institutional challenges.