T.J. Synnott and R.H. Kemp
T.J. SYNNOTT and R.H. KEMP are in the Unit of Tropical Silviculture, Commonwealth Forestry Institute, Oxford. Mr. Kemp heads the Unit.
The merits of natural regeneration, enrichment planting conversion and agri-silviculture in tropical forestry are relative and depend upon factors ranging from the ecology to economy
A fundamental feature of tropical moist forest ecosystems is their complexity and while this is a major problem for the manager it is a major strength of the system, in regard both to environmental hazards and to changing demands. The need to simplify and refine the system for wood production must therefore be set against the desirability of retaining the wealth of variability that may be required to adapt to future changes. It may be necessary, however, and is certainly possible for these two largely incompatible processes to be carried out separately, in different parts of the forest.
In most areas of tropical moist forest, natural regeneration has been relied on to provide the future crop. The silvicultural techniques have been intended to increase the stocking and growth rates of seedlings of valuable species, but techniques for inducing regeneration of chosen species have often been unreliable. In practice, it has only proved possible to increase directly the stocking of valuable seedlings if silvicultural operations can be timed to coincide with abundant seed-fall of valuable species. More success has been achieved with operations which increase the survival rates and sometimes the growth rates of existing valuable young trees by reducing competition from unwanted trees. Some techniques improve the proportion of valuable trees in a stand by eliminating unwanted trees without necessarily increasing their growth rates of seedling numbers.
Formerly, uniform systems (especially those termed shelterwood systems) were designed to achieve abundant regeneration of valuable species by careful manipulation of the canopy. However, complicated systems often proved difficult to organize, as well as being unreliable or unnecessary. In current practice, uniform systems aim to ensure the survival and growth of an adequate amount of the existing regeneration. A fundamental feature of these systems is that selection is operated against the unwanted trees over the whole regeneration area, irrespective of the local presence of valuable individuals. If there are few desirable species, uniform systems are likely to fail. However, the extensive destruction of much of the stand and the canopy caused by the more intensive uniform systems, usually results in relatively abundant regrowth of seedlings of certain light demanding or pioneer species. If these species produce useful timber, supplementing the existing valuable trees (e.g., certain Dipterocarps in Malaysia) or if a market subsequently develops for them (e.g., Mansonia and Triplochiton in Ghana, Maesopsis and Funtumia in Uganda) the systems may subsequently be judged successful.
Dawkins (1959) has stressed that the timber yield from most forms of tropical moist forests is likely to decline under a polycyclic system (including selection or stratified uniform systems) because of the extensive damage to young regeneration caused by felling large-crowned, upper canopy trees and the inability of most of the desired timber trees to grow vigorously when under the shade of older and larger trees. Further, in most areas of tropical moist forests, methods involving only partial canopy opening have not succeeded in inducing or increasing the regeneration of the most valuable species as much as required.
For these reasons, and because of the relative ease of administration and control of uniform systems, polycyclic systems are not widely practiced. In some cases where such a system is advocated or practiced, as in the Mora forests of Trinidad (Bell, 1971), in the Philippines and in Queensland or perhaps in Malaysia, natural regeneration of valuable species exists in larger numbers and with more even size-class distribution than is common in tropical moist forests in other countries. The stocking of valuable seedlings is in most cases reduced rather than increased by harvesting and refining operations, but may remain adequate.
The Ghana Selection System (called a selective system by Nwoboshi, 1975) includes tending operations designed to increase the growth rates and survival of young individuals of valuable species larger than 10 cm diameter which exist at the time of harvesting. Britwum (1975) states that the canopy disturbance induces some new regeneration. The abundance and species composition of the seedling regeneration cannot be precisely controlled although it can be predicted within wide and general limits. The growth rates and stocking are likely to be lower after slight canopy disturbance than after the more drastic disturbance involved in uniform systems, although the regrowth in both cases is likely to include a majority of shade-intolerant species.
Most tropical countries have used nursery-raised planting stock to improve the species composition or productivity of indigenous forests, often developing many systems in practice over many years and large areas. General principles and local experiences are summarized by FAO papers (1970, 1974), Lamb (1969), King (1968), Nwoboshi (1975), Moore (1975) and others including the "principles and technical guides" which must be followed if line planting is to succeed. Many typical methods have been described, and many intermediate examples are known, differing mainly in the planting density and the extent to which the existing young valuable trees, or later regrowth, are to be retained for harvesting. They may be loosely grouped as follows:
I. Enrichment planting in gaps, when trees are planted only on spots where no young valuable trees exist, so that the future crop will include the unharvested young valuable trees and subsequent regrowth of valuable species, supplemented to a variable extent by the planted trees.II. Line or group planting, when trees are planted in cleared lines or in more or less regularly spaced groups in a matrix of partially cleared forest. The intensity of planting may vary from a minor enrichment, supplementing natural regeneration, to a stocking sufficient for a full final crop, amounting to a conversion planting. In practice, a proportion of natural regrowth is usually accepted in the development stand.
III. Close planting, when enough trees are planted to provide at least the number expected in the final crop, without any contribution from natural regrowth. They may be planted at a closer spacing to allow for losses, selection and thinning. The pre-existing forest and regrowth is either cleared progressively to make room for the planted trees or it is completely cleared before planting.
The problems of clearing forest for enrichment or conversion have resulted in the development of many systems involving arboricides, charcoal burning, bulldozers and agri-silviculture. The various systems and field techniques of crops establishment have varying advantages and disadvantages, with different relevance in different regions, according to such conditions as the availability and skill of labour, the suitability of terrain and soil for mechanical equipment, the availability of markets for the produce, competing demands for land, and so on.
Although prediction of market demands and opportunities for the final crop must be uncertain at the time of regeneration of that crop, an objective or objectives must be set and priorities be clearly established between them. This uncertainty places a premium on flexibility in management, to accommodate changing demands, and this consideration may in turn influence the choice of regeneration method. The more exactly the method is designed to meet a special market requirement the fewer the options likely to be open for changes in management objectives later in the rotation. A major defect of most natural regeneration systems is the inability to predict exact production levels either of particular species or classes of timber or indeed of total merchantable wood. However the natural variability of the forest is likely to assist flexibility to accommodate changing markets.
A continuing and increasing world demand for wood can be predicted with confidence and most tropical countries, with rapidly increasing populations and rising living standards, can expect increased local demand for all classes of wood products. A major objective may therefore be the maximum production of useful timber from a limited area of forest. In Nigeria, for example, it has been calculated that, in order to meet the country's future wood requirements, it will be necessary to convert the whole high forest estate of 1½ million hectares to intensive production within 50 years. In these circumstances, and on present knowledge, the operation of a natural regeneration system is not sufficiently productive (Lowe, 1975a). It is necessary to practise the maximum interference, through enrichment or conversion to plantations. Volume production in naturally regenerating tropical moist forests of mixed species is frequently of the order of 2 cubic metres per hectare per year whereas plantations of fast growing species can readily achieve ten times this rate of production. Considerable increases in the final crop can be achieved by line-planting at little more than the expected final stocking in partially cleared forest (e.g., Aucoumea in Gabon). Close planting after complete clearance of the forest may provide valuable intermediate returns from thinnings, which may double the total yield, when compared with plantations raised to maturity at the final spacing, without thinning. Even higher rates of production may subsequently be achieved in plantations through the control that may be exercised over selection of species, provenances and individual trees and the production of improved material through breeding.
UPLAND DEVELOPMENT IN INDONESIA when in doubt, choose the natural system
Up till now, almost all the valuable tropical hardwood timbers have been harvested from naturally regenerated forest and this is likely to be so for the rest of the present century. In many areas it has not been possible to increase or even maintain the stocking of the currently commercial species and if they continue in demand they are likely to increase in value as the supplies diminish. Countries possessing tropical moist forests of valuable hardwoods, therefore, may have a great advantage in future world markets, provided that they can achieve their regeneration. Although a few major species, such as teak, are readily raised and managed in plantations many are not, either because of the dangers of insect damage in concentrated populations, or for reasons such as slow initial growth rate, very short seed life, susceptibility to exposure or other disturbance during the nursery or planting phase, and other reasons. At the same time there are often difficulties in obtaining natural regeneration of such valuable species with any certainty. Nevertheless the stocking of valuable young trees before and after harvesting in the forest may be readily assessed and, if it is adequate to provide a final crop (e.g., a stocking of about 100 established trees per hectare) there is a clear case of retaining the natural regeneration, even though knowledge of the probable growth rates, and of possible ways of influencing the yield, may be uncertain.
Some other special classes of timber, such as long fibre for industrial use, may demand a greater degree of interference with the forest, to ensure an adequate supply of relatively uniform material, at a given age, with a greater measure of control over the growth rates and final yield. The conversion of the forest to managed plantations is then clearly indicated.
The influence of forest vegetation on soil and water resources in the humid tropics may vary greatly with the structure of the forest and the amount of disturbance during logging and regeneration. For this reason the method of regeneration must take account of the need to protect the stability of the system in areas where disturbance may cause accelerated soil erosion, unfavourable changes in stream flow or loss of soil fertility. The main dangers are the loss of topsoil, sedimentation of streams, canals and reservoirs, the increased magnitude of flood flows, a reduction in water yield during low flows, and the loss of soil nutrients through accelerated leaching and the disturbance of soil nutrient cycles. Although very little experimental work has been done to quantify these changes there is sufficient evidence of the effects of forest clearance to show they can be severely destructive of a country's water resources (e.g., Daniel and Kulasingam, 1975).
Logging inevitably disturbs the forest structure and leads to severe local soil disturbance, although careful location of roads and tracks can reduce the damage (Gilmour, 1971). The degree of soil erosion and of reduction in water infiltration depends not only on the slope, soil structure, soil depth and rainfall intensity but also on the size of the area affected and the time taken to restore a vegetative cover. Complete clearance of forest, to create plantations, causes maximum exposure, at least temporarily, to sun and rain and may also entail burning and soil compaction. Although the plantations may later attain much of the stability of the original rainforest in regard to nutrient cycles the type of forest canopy may be very different, particularly if the crop is deciduous and composed of a single species. Wherever soil and water resources may be vulnerable to changes in forest structure the retention of an evergreen effective cover by the use of natural regeneration or enrichment methods may be preferred. If retention of an effective forest cover is the primary objective it can be most simply achieved in most areas of tropical moist forest by unaided natural regeneration.
Any operations to exploit and regenerate the forest will have some impact on the genetic resources of both flora and fauna. Tree species which are major constituents of climax forest, do not regenerate adequately after logging, and are not readily cultivated on present knowledge, are clearly in danger of depletion or even extinction, depending on the degree and extent of interference. The present fragmentary knowledge of the ecology and genetics of tropical hardwood species in tropical moist forests illustrates the interdependence of flora and fauna, and the need to conserve representative samples of the major forest types, as "strict natural reserves" or "virgin jungle reserves." The minimum effective size for such areas to act as gene pools for long-term conservation is not known on such reserves, or even their dereservation, unless they are buffered by a surround of managed forest. In such a buffer zone a system of natural regeneration, perhaps assisted by some enrichment, would clearly have great advantages in rendering the central reserve more effective, and at the same time extend the area of protection for many of the species. Conversely more intensive systems, such as conversion planting and agri-silviculture, greatly reduce the genetic diversity, perhaps in favour of exotic species.
The other major services to society which may be influenced by the choice of regeneration method are the provision of recreation facilities and the protection of the human environment, particularly in regard to health hazards.
Wherever the provision of employment in rural areas is an important social objective the techniques employed in regeneration can be chosen to be labour intensive. The more extensive systems, such as natural regeneration or enrichment, involving less interference with the forest, may be more readily adapted in this way than the more intensive systems, in which the use of mechanical equipment for clearing, planting and weeding may have operational advantages. Nevertheless where an adequate labour force is available for such intensive operations to be carried out on schedule the use of manual labour may reduce the dangers of adverse environmental impacts, such as unnecessary soil disturbance or soil compaction, which might be associated with the use of mechanical equipment. If the objective is to provide continued work for a settled community then the more intensive systems, such as conversion planting and agri-silviculture, may be preferred, since they provide a greater concentration of jobs and greater possibility for employment on work such as thinning and tending, at intermediate stages before final harvesting.
The extent to which the regeneration of forest by agri-silviculture may also contribute to food production is illustrated by the Nigerian experience, where 10000 hectares are cleared annually for combined arable cropping and timber crop establishment, producing food crops to a value of over U.S.$5 million (Lowe, 1975b). In many areas of tropical moist forest the soils are vulnerable to leaching and erosion and incapable of sustained arable farming without the intervention of tree fallows. The combination of arable farming and timber crops offers considerable scope for sustained rural communities and appears more profitable than either separately.
At present these are minor objectives in most areas of tropical moist forest and unlikely to influence greatly the choice of regeneration systems. The ease of access, aesthetic appearance and the influence of the forest on wildlife vary at different stages in the life of the crop, and their value varies according to individual human judgements. In some circumstances the way in which the type of forest developed may effect populations of parasites or vectors of human diseases could be an important consideration. However this may more properly be treated as a constraint on the use of the forest than as an objective.
The choice of regeneration method has an important influence not only on the type of forest produced and its productivity but also on the way in which the nation's resources of land, forest vegetation, staff and finances are used to achieve the objective. When resources are severely restricted there may be strong social and political pressure for them to be used in ways that are evidently profitable within a short time; short, that is, in comparison with a timber rotation period. This pressure has an important and sometimes decisive effect on the choice of regeneration method.
The greater the area of land available for forest production the less the pressure for intensive use of the land. However, when the market demands a high rate of production of wood from a limited area, or when other forms of land use are competing strongly for forest land, natural regeneration systems are disadvantageous compared with the more intensive systems, particularly conversion planting and agri-silviculture. The competition may come from shifting agriculture, when farmers wish to move from soils already degraded by agricultural use, to the soils kept fertile by forest. Although natural regeneration of the forest might in the long term achieve maximum benefit from limited resources the choice of conversion planting or agri-silviculture may then be made as the only way to resist the pressures for dereservation of the land. With rapidly increasing human populations this factor is bound to exert increasing influence on the choice of regeneration methods.
Clearance of forest for farming has been an important influence on the occurrence and structure of tropical moist forests throughout the tropics and over a very long period. The most fertile soils, capable of sustained arable cropping, have become the base for settled communities and in general it is the poorest soils which have been allowed to remain longest as forest, since they have not permitted the development of a stable agricultural system (Van Baren, 1974; Fraser, 1975). The low level of human population in such areas leads to a low level of pressure on the land for other uses than forestry and also limited availability of labour for any intensive operations. In these circumstances natural regeneration systems, which aim to retain a considerable part of the original forest complexity, with a mini mum of disturbance compared to other systems, are least likely to upset the balance of the nutrient cycle. Accepting a low level of productivity imposed by the limitations of soil and climate the use of a system which demands only low levels of financial and labour resources can be an advantage.
Low productivity of many areas of tropical moist forest is due not only to the limited capability of the site but also to the fact that there are very few valuable species present, that their regeneration is inadequate, and that their rates of growth are comparatively low. In these circumstances the introduction of a greater number of valuable trees, perhaps with faster growth, is clearly advantageous, provided that the anticipated increase in value of the crop will bear the cost of establishment at the end of the rotation (perhaps with some contribution from intermediate yields). The use of selection and breeding methods may offer further possibilities for increasing the productivity of the forest in later rotations. By contrast the possibilities for increasing the productivity by use of natural regeneration alone are very limited. A related factor that may be taken into account is the effect of competing vegetation, such as climbers. These may suppress or distort regeneration of valuable species from natural regeneration or enrichment methods, as well as more intensive plantations. Although the cost of weeding and climber control may be more readily borne by the more intensive plantation crops, which have a higher value per unit area, the emergence of dense growth of weed species in dense uniform conversion planting has been a severe problem in some areas (Brazil, 1975).
In many developing countries the lack of capital to invest in long-term projects such as timber production, where financial returns are relatively low and long-deferred, is more severely limiting to the choice of regeneration methods than is the land availability. Natural regeneration is relatively cheap and an equal investment can be spread over a much wider area, thus retaining, perhaps, a larger forest estate, with an assurance of continued production of wood even if the growth rates and future market value may be uncertain. If unwanted trees can be cleared cheaply for charcoal production the cost of enrichment and selection weeding may be so low that a more predictable crop of higher value may be assured at a rate of return that compares favourably with more intensive methods such as conversion planting, despite the higher yields achieved by the latter method. However if the forest is to be used as the basis for development of wood using industries in the area a greater degree of predictability in the quality and yield, and a greater intensity of production per unit area, may be required. In such a case the conversion of the forest to dense, high yielding plantations may be possible, perhaps using private capital of the companies concerned in the initial investment.
The initial cost, through its effect on the discounted value of the final crop, is a major factor influencing the choice of regeneration method and it is influenced in turn by the present low value of most of the timber in tropical moist forests. Nevertheless there is a high price differential between the more valuable hardwoods in the forests and the fast grown plantation crops which may replace them. In some cases, as in Queensland, Australia and in Trinidad, for example, this difference in value may compensate for the longer rotations and lower yields per unit area. A slightly higher market price for present commercial species, and acceptance by the international market of some presently non-commercial species, could improve the financial attraction of natural regeneration and enrichment methods in comparison with intensive plantations of fast growing species of lower unit value.
Agri-silviculture, as pointed out by Lowe (1975b), assists not only in amortizing the costs of establishment of the tree crop, but also in achieving greater financial profitability than either arable farming or timber crops separately. However he also indicates the problems that arise in operating such a system, and the dangers if there is inadequate control of the system in practice.
Any method of regeneration in tropical moist forests presents technical and operational problems related to the complexity of the ecosystems, with many species and site differences over small areas, the difficulties of access and movement in the forest and the frequently large extent of the area to be treated. In many developing countries there is a shortage of trained staff and sometimes of labour available for work in the relatively remote and often uncomfortable conditions of the forest. The more extensive systems of regeneration, if they also demand close attention to the distribution, composition and status of the regeneration, whether individuals or groups of young trees, present the greatest problem in ensuring adequate control and supervision. The administrative difficulties are greatest if frequent visits need to be made to each area of forest for a sequence of relatively minor operations, as in the original Tropical Shelterwood System developed in Nigeria.
The more intensive systems avoid some of the problems but create others if they involve repeated tending operations and lead to later thinning, pruning and protection from pests, fire or other dangers. The more intensive the system, and the greater the initial investment, the greater the potential losses from failures in control and supervision. Nevertheless there is in general a better understanding of the techniques involved in the management of artificially regenerated forests, particularly close planted stands, than those involved in the management of naturally regenerated tropical moist forests. Assuming that continued research may improve our knowledge and understanding of the dynamics of the natural system, and that education and training programmes will produce more skilled staff for future management needs, it may be preferable to concentrate the presently available resources on to the intensive regeneration of a relatively limited area of forests, provided that in so doing we do not increase the danger of dereservation of other areas through apparent lack of management.
Whenever there is some doubt concerning the choice between natural regeneration and more intensive methods then the greater robustness and long-term security of the natural system should be taken into account, and the benefit of the doubt should be given in favour of maintaining the natural forest until the case for other forms of management is better proven.
