Contents - Previous - Next

Appendix V


1. Soil Erosion

The potential erosive action of logging, log skidding, logging-road construction and other forestry activities is well known, and an abundant literature exists concerning its control in various geographical settings (see Bibliography, Annex B). The adoption of a particular set of methods of logging, log yarding and forest-road construction designed to minimize soil erosion depends on many variables, such as climate, typo of forest, terrain, geology, soils and manpower and skills available. Rigid guidelines for erosion control of universal application are, therefore, not feasible. The following list is only a reminder of common methods of soil erosion control or prevention in forestry.

- Avoid clear-cutting if at all feasible.

- Adopt "small coupe logging" that results in a checkerboard pattern of logged (5-25 ha) and unlogged areas (benefits offset by greater length of secondary logging roads).

- Leave forest screens (buffer zones, sediment filters) 20-40 m wide along streams.

- Avoid logging steepest slopes; avoid steep roads (1 in 3 maximum in the humid tropics?).

- Avoid logging or construction on highly erosive soils (commonly those derived from granites and sandstones).

-Build roads, trails and yards to the minimum dimensions required for efficient operation.

-Plan felling so as to minimize log skidding.

-Avoid building temporary logging roads every logging season.

-Build roads along ridges, as opposed to mid-slopes or valley bottom.

-Avoid skidding logs or building spur roads perpendicular to main slopes.

-Introduce occasional level or negative grade in logging roads so as to allow them to drain.

- Drain and stabilize minor roads and skid trails as soon as possible after use; ideally, build roads and trails that effectively self-destruct after use.

- Avoid using earthfill dams across streams as temporary bridges.

- Stabilize road outs and other bare areas with mulch, wood chips or sod; use sound engineering design to avoid slumping of road outs.

- Minimize or avoid use of burrow pits, which can be important sources of sediment.

- Use sediment traps (straw-bale or brush barriers, plastic-sheeted stilling basins) in small stream.

- Schedule logging so as to avoid periods of high-intensity rainfall.

- Use cable yarding, if possible, and if the manpower is available to use this method correctly.

- Use "soft technology" (eg., "manual" selective cutting, offset by high labour demand, low productivity and relative high accident rate).

- Reforest as soon as possible, if necessary by using shortcuts such as aerial seeding, vegetative propagation, or planting fast-growing intermediate tree crops (eg., Leucaena leucocephala).

-Treat compacted, runoff-producing areas with mulches or clasp rooted annuals or biannuals (eg., legumes).

- Terrace or use grass berms, palisades or wattling-and-staking on the steepest, most erosion-prone slopes.

- Use several small landings or yarding areas instead of few large ones.

- Use corduroy in sensitive areas, such as along streams or in unavoidable stream crossings.

- Install permanent culverts under permanent roads; design drainage for low-frequency storm runoff.

- Repair road washouts as soon as possible; in general, keep roads well maintained.

- Keep skid trails at least several hundred metres apart, ensure that drainage from one trail is not added to that of another trail.

- Skid logs with one end off the ground; avoid skidding in streams.

- Build berms around large yarding areas in order to contain sediment.

- Log uphill if at all possible ("high lead").

2. Soil Deterioration

Tropical soils are generally fragile and of low inherent fertility. Exceptions are certain soils of regent volcanic origin, which can be stable and fertile. Owing to the low fertility of the mineral portion of the soils, tropical moist forest in particular perpetuates itself by means of a relatively closed cycling of nutrients from living biomass to litter to living biomass. This cycling is critically assisted by saprophytes that shortcircuit complete mineralization of organic matter by transferring nutrients still in bound form to plant roots. Many tree species of the tropical moist forest (Dipterocarpaceae, some Tiliaceae, Myrtaceae, Sterculiaceae; Unesco/UNEP/FAO, 1978) are apparently mycorrhizal; mycorrhizae are particularly important in infertile soils, where they may be the principal determinants of species composition at the needling stage. In addition, the cation-exchange capacity of tropical soils resides principally in the organic layer, as opposed to the clay component in the mineral horizons, as is the case in the temperate zone.

Logging and related micro-climatic alteration and soil erosion can permanently disrupt tropical forest ecosystems through the loss of organic matter, deep leaching of nutrients beyond rooting depths and alteration of the microflora and fauna. Under extreme conditions, irreversible "tropical deserts" can be created, primarily where long-term leaching had already left Pew nutrients in the mineral portion of the soil (so-galled "white sands"), and most of the cation-exchange capacity and available nutrients are lost once the litter is removed.

Owing to the nature of soils and of the total forest ecosystem in the humid tropics, mitigation of soil deterioration must start from the premise that clear-cutting should be avoided if at all practical. This is particularly true on ancient, highly leached soils (primarily soils that are not of recent volcanic or alluvial origin) and where the management objective is the natural regeneration of the high forest.

In addition to outright avoidance of clear-cutting, other mitigative measures include:

- Minimize clear-cut areas (strips in the 50-200 m range1) so as to avoid drastic soil losses and micro-climatic changes, to ensure the proximity of seed sources and to allow the adjacent forest to encroach as it sheds its litter over the logged area.

1 depending on elope gradient, exposure to wind, etc.

- Cover logged areas with logging debris, mulches or with cover crops (eg., herbaceous legumes; trees such as Leucaena) that exclude weeds and protect the soil but that do not interfere unduly with the germination and seedling growth of tree species.

- Experiment with mycorrhizal inoculations of bare logged areas by using infected litter from adjacent residual forest.

With selective cutting, the risk of serious soil deterioration is reduced but not necessarily eliminated, particularly if the forest is thinned out considerably. In East Kalimantan the removal of only 25 trees per ha has resulted in 30 percent of the total logged area having soils disturbed and compacted by tractors (Kartawinata, 1978).

3. Poor Forest Regeneration and Weed Invasion

Regeneration of tropical moist forest faithful to its original species composition is difficult, not only because of soil degradation and other environmental changes caused by logging, but also because this species-rich forest naturally tends to "mosaic" regeneration. Regeneration that is highly variable in space and time is particularly true of high forest growing in undifferentiated lowland terrain. Regeneration is highly variable because of the many species present and potentially able to occupy a vacant area, because of the periodicity of reproduction and such vagaries as the width of the gap in the canopy, density of seed trees in a particular area (usually low for most species), relative efficiency of seed dispersal, existing residual seedlings and saplings in the gap area and germination and seedling survival. Recent research also shows that some secondary forest species have an advantage over primary species by having dormant seeds stored in the soil for as long as years and more abundant seed (Cheke et al., 1979).

If, for the reasons stated above, natural regeneration is hard to predict in forest gaps caused by natural factors, the impact of selective logging on subsequent species composition is even more difficult to forecast. However? it appears that selective logging at a rate not exceeding 5 trees per ha has minimal impact on the residual vegetation, primarily temporary invasion of the gaps by weeds (eg. Eupatorium odoratum in SE Asia). On the other hand, tractor removal of only 25 trees per ha hen resulted in 30 percent bare ground and subsequent invasion by persistent woody and non-woody species), at least in SE Asia (Kartawinata, 1978). Among these species were Macaranga spp., Trema spp., Anthocephalus chinensis and Imperata cylindrica; Anthocephalus has remained dominant on such disturbed soils for over 40 years (Kartawinata, 1978). Few commercial dipterocarps were found to colonize such soils. Anthocephalus can, however, be a useful timber species in its own right. In Sabah, "normal" forest was found more than 40 years after selective logging on those soils that had not been disturbed by the logging (Kartawinata, 1978).

Regeneration of selectively cut forest is also made difficult by mechanical damage to residual treas. In one case, as much as one-half of the residual stand was found damaged; elsewhere, only 35 percent of the remaining basal area was left undamaged despite a removal of only 10 percent of the total b.a. (Hanzah, 1978). In addition, single seed trees left after logging usually die from "isolation shock", that is from changes in the total bioclimatic environment (Suparto et al., 1978; Unesco/UNEP/FAO, 1978).

The examples given so far are only intended to illustrate some of the impacts of logging on vegetation that have been documented so far. There are, of course, many regional differences in forest regeneration owing to different floras and environmental conditions. In Papua New Guinea, for example, selective cutting of Araucaria cunninghamii means drastic changes in vegetation and in soil nutrient status. Thus, whereas N and organic C return to former levels rapidly, Ca and K take much longer. Two stages of secondary vegetation are apparently needed before Araucaria finds nutrient, litter and light conditions. suitable for regeneration. Araucaria forest requires even more careful selective logging than Malaysian, Philippine or Indonesian dipterocarp forests in order to ensure successful regeneration (Burgess, 1971; Cheah, 1978; Enright, 1978; Rapera, 1978; Suparto et al., 1978). In the Philippines, forest management is difficult enough as it is. Thus in that country, about 60 percent of the trees 20-70 cm in diameter are left as seed trees, and yet even at this rate there is no guarantee that the regeneration of dipterocarp forest will occur (Rapera, 1978).

Tree age, site and other considerations must be taken into account in order to ensure proper regeneration. Dipterocarps left as seed trees must be sufficiently mature to perform this function: for example, some Indonesian dipterocarps less than 1 m in girth, that is less than 30 years old, are still sexually immature (Kartawinata, 1978). Especially poor regeneration can be expected on swampy and sandy soils (Sudiono and Daryadi, 1978). In some cases, efforts at regeneration may have to resort to vegetative propagation of timber species), in order to by-pass the critical germination-seedling stage (Momose, 1978). Recommendations for ensuring the regeneration of Malaysian dipterocarps are given by Nin (1978).

Even the brief review of the subject given so far shows that the regeneration of tropical forests and the prevention or control of weed invasions are complex matters about which insufficient information is available today. The examples of impact show, however, that poor regeneration and weed invasions are problems that generally increase in proportion to the increase in the rate of selective removal and the use of heavy machinery. Judging from some SE Asian examples, it seems possible, however, that even upwards of 25 trees per ha can be removed without jeopardizing useful regeneration provided essentially manual methods of selective cutting are used. Elsewhere, it may be that the systematic planting of cover crops (eg., Leucaena) in logged areas can be used not only to ensure forest regeneration and to pre-empt the invasions of persistent woody and non-woody weeds, but also to provide useful products in the meantime.

4. Disturbance of the Forest Fauna

Systematic studies of the short- and long-term effects of logging on forest animals (so far, nearly all mammals, with an emphasis on primates) are only beginning to be made now, notably in Indonesia. No such studies were as yet available as late as the mid 1970s (Whitmore, 1975). As a result, few specific recommendations for protecting the fauna prior to, during and after logging can be formulated now, least of all for the various biogeographic provinces of the tropics and subtropics.

If little can be said about the larger mammals, even less is known about the safeguarding of the smaller mammals, other vertebrates and let alone the invertebrates in logged forests. In the case of the invertebrates, thousands if not millions of species), are probably still unknown to science (Myers, 1979a). Size has, of course, little to do with the intrisinc ecological importance of the species), the importance of the soil microfauna and of insect pollinators to forest ecology need only be recalled. These Guidelines can only emphasize the importance of considering all species in the protection of the environment, and of the need to pursue further research on all components of tropical ecosystems. Powerful arguments for preserving species diversity in the tropics and elsewhere are made by Myers (1979a).

Some tentative conclusions concerning the impact of logging on forest animals, mostly mammals, have been reached for parts of SE Asia. Arboreal mammals seen to be the most affected. Browsing animals are less affected; some, notably Bovidae and elephants, may be favoured by new browse. Some arboreal animals can be displaced to nearby high forest, but crowding stress can result if the displaced animals are added to resident populations. Thus, crowed orang-utans have been known to suffer reproductive stress, and birds, especially canopy dwellers, have displayed agitated and erratic behaviour after displacement (Whitmore, 1975).

Most birds of the dark tropical high forest are thought to be intolerant of secondary growth, primarily of its brighter light (Myers, 1979a). Some bird species), can be easily eliminated: for example, hornbills that need large hollow trees for nesting, and that may need as much as 10,000 km2 of forest in order for an adequate breeding stock to be present (Whitmore, 1975). In general, the fauna of the primary lowland moist high forest is unable to occupy the large new areas of secondary forest and associated scrub and grasslands (Whitmore, 1975). Exceptions are rodents that thrive in grasslands and those birds that find more edible fruit in secondary forest.

On the other hand, selective logging in peninsular Malaysia did not affect at least two primates (siamang and white-handed gibbon) (Whitmore, 1975). Similarly, in East Kalimantan and in Sumatra, Indonesia, no significant impact on primates was detected in selectively logged forest (8 or fewer trees removed per ha) after one year (Wilson and Wilson, 1978). In contrast, even light selective cutting could affect the Bornean gibbon if too many broad-canopied emergents are removed, as these trees are the gibbon's preferred sleeping habitat (Wilson and Wilson, 1978). The orang-utan is also potentially sensitive to selective logging because of its low mobility and its habit of escaping detection through immobility (Wilson and Wilson, 1978).

Local extinction of some species could occur if their highly specialized habitats are destroyed. One example is the proboscis monkey which feeds on few species of the mangrove forest (Wilson and Wilson, 1978). In general, however, it is believed that many species), can survive selective cutting at an intensity not exceeding 8-12 trees per ha, provided that the residual vegetation is not extensively damaged (Wilson and Wilson, 1978). It must be remembered that, in the long run, noise, human presence and excessive or illegal hunting associated with logging may prove more damaging to the fauna than the actual logging.

It is unrealistic to expect that environmental planning for single forestry projects can deal effectively with a matter as complex as the conservation of the fauna, particularly where the species), to be protected are mobile or migratory. Conservation evidently requires concerted efforts at the regional or national scales, including the establishment of parks or reserves. However, the present Guidelines recommend the following minimum steps for preventing or reducing the potential damage to the fauna at the scale of the individual forestry project:

- The environmental assessor of the forestry project should ascertain the presence of rare, endangered or protected animal species), in the general project area with the aid of IUCN Red Data Books or national lists (eg., Hardjosentono, 1978);

- The assessor should consult with local or regional conservation authorities or other wildlife experts in order to establish minimum measures for protecting the fauna (eg., identification of key habitats to be avoided; capture and resettlement of some animals; severing of lianas and other climbers prior to felling so as to reduce logging damages by entrainment; role of game officers before, during and after logging, briefing of forestry workers);

- The assessor should review logging and road construction plans with a view to avoiding important habitats or critical periods in the life cycles of certain species;

- In consultation with conservation authorities, the assessor should devise contingency plans and procedures, including individual responsibilities, for dealing with unexpected problems with animals during logging, such as the discovery of healthy or injured rare animals;

- The assessor should devise plans for controlling movement on logging roads, particularly if the project is within the buffer zone of a park or nature reserve.

Ultimately, there is no substitute for planning forestry projects on the basis of detailed ecological maps compiled from scientific field research. In the absence of site-specific data on the fauna and its ecology, it is recommended that a rough-and-ready classification of forest types or habitats in the project area be compiled from aerial photography and ground reconnaissance, ideally in consultation with wildlife experts familiar with the region. Potentially sensitive habitats should then be identified with the help of general guides on the fauna likely to be present. For example, if the orang-utan is known to be present? it would be prudent to identify concentrations of durian trees (Durio spp.) along the edge of a swamp as a potentially important habitat, given the feeding and other preferences of this primate.

5. Reduction of Plant Species Diversity and Loss et Genetic Resources

As in the case of the fauna, the conservation of plant species diversity and of genetic resources cannot be achieved primarily through the environmental assessment and impact mitigation of single forestry projects. The evaluation of the species diversity of forests, of the range of endemic species and of the minimum viable gene pool of tropical forest trees is probably a task even more complex than the conservation of, the fauna. This task is complex because of the many species present, many still unknown to science, small species populations and poor accessibility. Genetic work is particularly difficult because many species have unknown populations and genotypic variation life cycles and reproductive strategies that are poorly understood and, at a more practical level, reproductive organs that are often at poorly accessible canopy levels.

Again, the present Guidelines can only recommend minimum measures for ensuring that some damage is avoided, and that forestry projects are undertaken with at least some perspective on the needs of conservation. As in the case of the fauna and its habitats, the goal is eventually to be able to plan forestry projects on the basis of inventories and classifications of forest resources, of legal zoning of these resources and of strict reservation of those botanical resources that are deemed essential from a broad scientific and social point of view (IUCN/UNEP/WWF, 1980; Myers, 1979a, b; Poore, 1976; Unesco, 1974b; Unesco/UNEP/FAO, 1978). Much progress has been made recently along these lines in the Amazon basin and in Papua New Guinea. The lowland forests of Indonesia are currently the subject of coordinated study. In other regions or countries such as Peninsular Malaysia, Sarawak and the Ivory Coast, a fair amount of scientific information is available concerning the vegetation, but much remains to be done to translate this information into specific conservation plane to guide forestry activities.

The assessment of forestry projects should include the following minimum steps:

- Ascertain whether classifications of the forests for conservation purposes exist, including those that are tentative or unofficial. Examples of such classifications are those suggested by SUDAM for Brazilian Amazonia (Unesco/UNEP/FAO, 1978), for Southeast Asia (Whitmore, 1975/76) and for northern Australia and Papua dew Guinea (Specht, Roe and Boughton, 1974).

- Place the forestry project in the biogeographic context of the region or nation, using existing regional or national floras (see Bibliography, Annex B) and/or in consultation with national herbaria or other botanical research centres, with a view to assessing the relative scientific value of the project area (the area may be identified as containing rare species or unknown floras, as having high species diversity, as being an area of convergence of plant migration routes, etc.).

- Examine the project area in the light of generalizations concerning species and genetic diversity (see Unesco/UNEP/FAO, 1978, p. 187); species diversity is usually highest on average to somewhat infertile soils, whereas fertile soils usually have fewer species with larger individual populations (Bruenig, 1973; Unesco/UNEP/FAO, 1978; Whitmore, 1975); high diversity can also be expected in areas of geologic diversity and in unusual geologic situations, such as on outcrops of limestone or serpentine in areas of predominantly acidic rooks; other potential foci of high diversity are areas of convergence of plant migration routes (usually identified in regional or national floras or in floristic monographs on subcontinental areas) and so-called Pleistocene refugia of moist tropical vegetation (Myers, 1979a, b).

- Areas likely to support high species diversity as well as areas with largely unknown floras (other than timber species should at least be reconnoitred on the ground by a taxonomic specialist, who may be able to identify scientifically interesting areas to be avoided by logging.

- If unknown tree or other species are present, areas containing at least several hundred mature specimens of these species should be set aside; similar numbers of residual trees should be saved if in the course of logging unknown species are encountered; the present consensus is that such populations, at least in the humid tropics, may constitute a viable gene pool, assuming random gene frequencies; if gene frequencies are skewed because of the existence of subpopulations (as suggested, for example, by distinctive phenotypes in different habitats, or so-called eco-types), then several hundred mature specimens ideally should be preserved for each sub-population. Since in most oases little or nothing will be known of genotypic variation, a more practical approach wherever a new species is encountered might be to set aside populations that occur on the entire range of habitats (say, from valley floor to ridgetop) that the species seems to occupy in a particular region.

Contents - Previous - Next