by
Rudolf Heinrich
Forest Harvesting, Trade and Marketing Branch
Forest Products Division
Forestry Department, FAO, Rome, Italy
INTRODUCTION
World forestry is challenged by a number of issues such as the loss of the earth's biodiversity, forest decline because of air pollution and transformation of old growth forests in the temperate region, decrease of forest land due to conversion to other land uses in tropical countries, forest land degradation and tree stand impoverishment as well as generation of forest waste caused by inappropriate and unsustainable forest harvesting practices. Recently issues such as labelling of wood products, suggestions for trade restrictions and even boycotts of tropical timber from non-sustainably managed forests have emerged as further causes of concern in forest and wood products development.
Although great advances have been made during the last two decades in developing and introducing highly mechanised and specialised machinery in forest operations, which permit environmentally sound, economically profitable and socially acceptable forest uses to support sustainable forest development, there still is a great need to ensure the introduction and application of appropriate policies and practical codes of environmentally friendly harvesting practices with the aim to advance sustainability of both timber and non-timber forest products.
Worldwide in 1991, some 3.4 billion m3 of round wood (FAO 1993) have been removed from the world's forests of which a little bit less than half has been used for industrial purposes and the other half as fuelwood. The presently existing forest resources worldwide are estimated to amount to about 4.1 million ha. With an ever increasing rate of deforestation for other land uses (conversion to agricultural land, infrastructure, and urbanisation; presently the rate of forest decrease in the tropics alone amounts to some 15.4 million ha annually), it is evident that a concerted effort is needed to motivate policy makers, managers, technicians and forest operators to encourage forest development programmes that harmonise interests in conserving forests as well as to wisely use the potential of the forest while maintaining its full regeneration capacity.
HARVESTING AND SUSTAINABLE DEVELOPMENT
At the United Nations Conference on Environment and Development (UNCED) held in 1992 in Rio de Janeiro, forestry received major attention under Agenda 21, Chapter 11 entitled "Combating Deforestation". With respect to forest use, particular reference was made to the need to promote efficient utilisation and assessment to recover the full valuation of goods and services provided by forest lands and woodlands (UNCED 1992). In many forest operations, we can recognize that the full potential of forests and forest lands is far from being realised as a major source for development.
A prerequisite for sustainable forest utilisation is comprehensive pre-harvest planning, appropriate monitoring and execu-tion of operations as well post-har-vest evalu-ations, increasing the production of goods and services, particularly in broadening diversity of yield of forest use, covering timber and non-timber forest products. This should help to generate more income and employment; enhancing life of rural populations, without compromising the regenerative capacity of the forests and their continued contribution to human welfare, while satisfying the aspirations of goods and services for future generations.
There are numerous definitions of sustainability. In 1904, G.L. Hartig, Head of the Prussian Forest Administration in Berlin, defined sustainability as follows:
"Every wise forest director has to have evaluated the forest stands to utilise them to the greatest possible extent, but still in a way that future generations will have at least as much benefit as the living generation".
On the subject of sustainable development, in 1988 the FAO Council adopted the following definition:
"The management and conservation of the natural resource base and the orientation of technological and institutional change in such a manner as to ensure the attainment and continued satisfaction of human needs for present and future generations. Such sustainable development conserves land, water, plant and animal genetic resources and is environ-mentally non-degrading, technically appropriate, economical-ly viable and socially acceptable".
Perhaps the most widely quoted definition of sustainable development is that of the Brundtland Commission (WCED 1987): "development that meets the needs of the present without compromising the ability of future generations to meet their own needs".
In view of the dwindling resources due to forest decline in the temperate zone, conversion to other land uses and degradation in the tropical areas, it is essential that forest harvesting practices are carried out in a manner to guarantee the sustainability of the forest resource base.
It is well recognised that the complexity and diversity of the vegetative cover as well as the fauna of the various forests require well planned and controlled forest operations and interventions to make full use of the potential of wood and non-wood forest products, compatible with environmental conservation. In some instances, the optimal use of all forest products may not be feasible due to factors, which include environment, accessibility, availability of harvesting technology, resources, laws and regulations.
THE NEW DIMENSION OF HARVESTING
FAO's perception of forest harvesting in relation to socio-economic development has gone through a substantial evolution. In forestry, a concept which needs to be more frequently introduced is the use of a wider range of forest products based on sustainable resource management. In addition to timber, a multitude of non-wood products are available from the forests which can be most beneficial in terms of employment and income generation, particularly for the local population. In many instances, harvesting is no longer synonymous with logging as it deals with a wide array of non-wood outputs (de la Cruz 1989). A partial list of such products may include plants for ornamental and medicinal use; honey, resins, tannin, fruits, mushrooms, nuts, wildlife for food and hunting trophies. Harvesting may be defined as the procurement of raw materials from the forests. When recognizing this new approach, harvesting as an independent technical discipline has to be seen as an integrating function forming a strong link between the resources, forest-based enterprises and markets. FAO, besides its traditional work in wood harvesting, has recently developed an action programme on harvesting of non-wood forest products and has undertaken case studies on small-scale harvesting that includes increased involvement by the local population. This was published as FAO Forestry Paper No. 87 (de la Cruz 1989). A study was also made on the collection, processing and marketing of edible mushrooms from forest plantations in Chile (Donoso and Kilkki 1993). By 1990 the value of these mushrooms had reached about US$ 3 million annually.
PROGRAMMES AND PROJECTS OF THE FOREST HARVESTING AND TRANSPORT BRANCH
In responding to the urgent needs to enhance environmentally sound and sustainable forest harvesting practices worldwide and supporting the transfer of appropriate technology, FAO has created a network of communication among scientists and practitioners from industrialised and developing countries. The FAO Forest Harvesting Bulletin is published twice annually and allows for the sharing of information and experiences relating the new development in forest engineering, harvesting and transport.
THE FAO FOREST HARVESTING BULLETIN
The aim of the FAO Forest Harvesting Bulletin is to promote environmentally sound forest practices worldwide. The bulletin reports on FAO's programmes and activities in this field of specialisation, highlighting emerging issues and conflicts and reporting on new advances made to reduce the environmental impact of forest harvesting on forest stands, soil and terrain; improving timber utilisation; generating employment and income; preventing forest workers accidents and limiting health risks in forest operations.
The eight-page four-colour newsletter has now a worldwide distribution of nearly 4,000 copies per issue. The membership list of this network incorporates forest engineering specialists, forest planning and policy personnel, forest operations managers, technicians and forest workers; national and international civil servants interested in forestry matters and representatives of grass roots level movements.
Presently the Forest Harvesting Bulletin is sent to institutions such as government forest administrations, forest universities, technical schools, training centres, research institutes, forestry libraries, forest and forest industries enterprises, forest operation contractors, non-governmental organisations, FAO/UNDP Regional and Country Representations and to the Regional Development Banks and the World Bank.
The suggestion to establish a worldwide network in the field of forest engineering, harvesting and transport emanated from recommendations made at the expert consultation on the FAO Action Programme on Forest Harvesting Training held from 23-27 April 1990 in Kotka, Finland. The FAO Forest Harvesting Bulletin essentially contains an editorial, featuring forest harvesting issues viewed in a general context; reports on the application of new technologies in forest road planning, construction, harvesting and transport, activities undertaken by national or international organisations in particular by ILO, UNIDO, ISO, ITTO, UNEP, etc., as well as meetings related to engineering and harvesting.
PROGRAMME ON ENVIRONMENTALLY SOUND FOREST HARVESTING
The overall purpose of this programme is to contribute to sustainable development by devising, testing, and helping to implement improved technologies for timber harvesting in tropical forests. Timber harvesting is considered here to comprise the aggregation of all operations relating to the felling of trees and the extraction of their stems or other usable parts from the forest for subsequent processing into industrial products. The term harvesting technology refers to the application of scientific and engineering principles, in combination with education and training, to improve the application of labour, equipment, and operating methods in the harvesting of industrial timber.
Activities being developed under this programme stem directly from critically needed tasks as identified in the Rio Declaration and in Chapter 11 of Agenda 21. For instance, Principle 4 of the Rio Declaration notes that "In order to achieve sustainable develop-ment, environmental protection shall constitute an integral part of the development process and cannot be considered in isolation from it." Principle 11 says in part that "States shall enact effective environmental legislation" and Principle 17 reads "Environmental impact assessment, as a national instrument, shall be undertaken for proposed activities that are likely to have a significant adverse impact on the environ-ment and are subject to a decision of a competent national authority." Similarly, Chapter 11 of Agenda 21 states that "unsustainable commercial logging" must be halted, that it is essential to develop "environmentally sound methods and practices of forest harvesting which are ecologically sound and economically viable," and that forestry must contribute to both economic development and environmental protection by endeavouring to develop "efficient conversion technology and improved sustainable utilization of harvesting and process residues".
The major activities for the Programme on Environmentally Sound Harvesting Tech-nol-ogies to Sustain Tropical Forests, as currently being planned by the FAO Forest Har-vesting and Transport Branch under the Forest Products Division, are listed below. To the extent possible, these activities are being initiated under FAO's Regular Programme but external funding will be needed for their full implementation.
_ Develop comprehensive guidelines for the application of harvesting personnel, equipment, and techniques in forest operations. These guidelines will form a model "code of practice" that could be adopted by governments to reduce environ-mental impacts associated with industrial timber harvesting operations.
_ Conduct case studies comparing conventional and improved harvesting tech-nologies in each of the three principal regions of the tropics. These case studies will be used to demonstrate improved practices, to measure the benefits and costs of such operations, and to foster the transfer of environmentally sound technology to developing countries.
_ Reduce waste associated with forest operations by developing technologies to improve the utilization of forest residues.
_ Develop methodologies for considering interactions between wood and non-wood forest products in order to reduce the impact of timber harvesting on the growth, collection, and harvesting of non-wood forest products.
_ Develop and test procedures for making comprehensive environmental impact assessments of proposed harvesting operations in tropical forests.
_ Conduct training sessions and expert consultations with the objective of contribut-ing to the application of improved harvesting technologies by government agencies and by private companies and contractors in developing countries.
FOREST HARVESTING TRAINING, EXTENSION, AND EDUCATION PROGRAMME
In a survey of training needs carried out by FAO with the support of the Forestry Training Programme of Finland in 29 countries, it became evident that due to the temporary nature of forest operations as a seasonal work, there is a high turn over of forestry personnel.
This certainly will have a great impact on the needs of training facilities; types and curricula of training; extension, and educational programmes. In general, it was observed that the subject area forest engineering, harvesting and transport had a time share of 10-15% of the total curriculum at Forestry Universities.
At the technical level, reduced versions of the university curricula were taught. Training at the vocational level was the most needed. A considerable shortage of training facilities and appropriate training material has been noted.
The Forest Harvesting and Transport Branch with the assistance of various bilateral donor agencies, has dedicated a considerable effort to carry out various training courses, workshops, seminars and expert consultations on a broad range of topics covering planning and management of forest operations; health and safety issues in forestry work; the transfer and use of appropriate harvesting technology; mountain logging; forest road planning and construction; forest operations in the tropics; forest harvesting and environment.
A recent initiative under this programme is the Seminar on Economics and Management of Forest Operations for Countries in Transition to Market Economies, which was held in Ort/Gmunden, Austria from 27 June to 2 July 1994.
The objectives of the seminar are to provide participants with some basic information on the principles of economically efficient and environmentally sound forest harvesting and engineering operations in the context of a market oriented economic system.
Special efforts were devoted to highlight legal, institutional, organisational and administrative measures through government services and support programmes of the Boards of Agriculture and Forestry in order to enhance the development of private forest enterprises based on the Austrian experience.
Participants had an opportunity to share experience and information by presenting their country reports, highlighting the success, problems and difficulties encountered in transforming centrally planned enterprises into market oriented enterprises. In this context, information was provided on the countries' forestry situation, legislation, types and sizes of forestry enterprises, level and choice of technology, production systems, employment and income, environmental issues, forest extension, training, education and research.
EQUIPMENT INFORMATION DATA BASES
The FAO Forest Harvesting and Transport Branch has established a computerised data base on forest engineering, harvesting and transport machinery, equipment and tools of main manufacturing companies and suppliers throughout the whole world. Presently the system contains information on more than 240 companies worldwide.
The information available comprises a wide range of equipment; incorporating basic, intermediate and most advanced mechanised harvesting technology and systems. The aim of the data base is to provide more efficiently technical services in assisting governmental, non-governmental and private forest enterprises to select appropriate equipment and machinery in relation to their socio-economic needs and forestry conditions.
Based on the experience in various field projects, data have been also collected on productivity and costs of various types and systems in forest harvesting. As a result of this data base, rapid evaluations on productivity and costs in forest harvesting can be undertaken.
The establishment of these data bases has greatly increased the efficiency and capacity of technical backstopping in FAO member countries in this field of competence.
HARVESTING OF NON-WOOD FOREST PRODUCTS
In countries as diverse as Sweden and Japan it has been estimated that the economic value of non-wood forest products amounts to between 20 and 25% of the value of wood products. Non-wood forest products are even more important to the economies of many developing countries. A few well-known examples include Brazil nuts, rubber, palm hearts, mushrooms, honey and beeswax, rattan, and bamboo. These products, and many others which are not so well known but are often more important to local people, have the potential to enhance the value of standing forest and therefore reduce the likelihood of deforestation. Harvesting, transportation and storage techniques for non-wood forest products are often complicated because of short harvesting seasons in which timing and rapid delivery to markets can be critical. FAO is working with specialists in non-wood forest products to develop improved harvesting, processing and storage techniques in order to help maximize the contribution of these products to income and employment in rural communities.
PUBLICATIONS
The Forest Harvesting and Transport Branch has created a new series of special reports on Forest Harvesting Case Studies which deals with innovative systems and techniques to improve harvesting standards, productivity, costs, and reduce the environmental impacts of forest operations. The aim of these studies is to thoroughly document environmentally sound and sustainable forest utilisation technology which could be of interest to many forest enterprises in a number of countries.
Two recently published studies under this series dealt with reduction of wood waste and non-wood forest products. They are:
_ Reduction of wood waste by small-scale log production and conversion in tropical high forest in Papua New Guinea;
_ Cosecha de Hongos en la VII region de Chile.
Further studies expected to be published will deal with:
_ Use of elephants in wood-extraction in Sri Lanka;
_ Low impact harvesting in buffer zone forests of protection forests in Uganda.
_ Improved harvesting techniques in tropical forests in Brazil, Congo, Peru and the Philippines.
For training purposes, a few practical oriented manuals have been prepared. Two were published under the FAO Training Series: the Chainsaw Manual and the Design Manual on Basic Wood Harvesting Technology.
The aim of the design manual is to provide ideas for techniques and equipment that will improve productivity at little or no investment; encourage the use of local skills and materials, and promote and encourage the involvement of the local population in forest utilisation. The equipment suggested in this manual is simple and inexpensive to make; reduces work loads and thus energy demands of human labour, and can easily be made from locally available materials in modest blacksmith shops, maintenance facilities of local communities or at the site of the task at hand.
Two further training materials were prepared on Cable logging and forest road planning and construction, subjects which require a very special training effort due to the complicated nature of the activities.
Many publications of the FAO Forestry Department's series, the FAO Forestry Papers, have had a major focus on forest harvesting and transport in the 110 papers published in this series. Since its inception in 1977, 15 papers have been devoted largely to issues relating to forest harvesting and transport.
The series is aimed at scientists and educators; professional people working in forestry or related fields, and high level decision makers, whose judgements may influence the future course of forest conservation and development.
REFERENCES
De la Cruz, Virgilio. 1989. Small-Scale Harvesting Operations of Wood and Non-Wood Forest Products Involving Rural People. FAO Forestry Paper 87. Rome.
FAO. 1993. Yearbook of Forest Products 1991. Rome.
Donoso J., and Kilkki, R. 1993. Cosecha de Hongos en la VII Región de Chile. FAO, Rome.
UNCED. 1992. United Nations Conference on Environment and Development, final advanced version of Agenda 21, Chapter 11, Combating Deforestation.
WCED. 1987. Our Common Future. World Commission on Environment and Develop-ment, convened by the United Nations General Assembly, New York. Oxford University Press, Oxford, UK.
by
Rihko Haarlaa
Department of Forest Resource Management
University of Helsinki, Finland
Abstract
In Finland, sustainable forestry has been practised for centuries and environmental protection aspects have always been considered in forest management and wood harvesting practices. The mechanization of forest work has been fast during the past decades. More than 80% of manual wood cutting has been replaced by harvester operations. Wood transportation by horse was given up about 20 years ago; subsequently agricultural tractors and later the specially designed forest tractors, the forwarders, have been used for off-road wood transportation. The main reasons behind this development are the higher productivity and the lower costs resulting from the use of these new methods. However, eventual damages caused to nature as well as other environmental aspects are gaining increasing importance in the decision-making process when deciding on actions to be taken and machinery to be used in forest operations.
INTRODUCTION
Finland is a country where the forests have always played a central role in the life of the people. Sustainable forestry has been practised during the centuries to the point that the growing stock has increased so much, that today there is more wood in Finland than ever before. During the last 50 years, sound forest management practices have been systematically followed, unproductive sites have been made productive and the results from forest research have been applied, even in private forestry, thanks to the efficiency of Finnish extension organizations. At the same time, issues like saving nature's biodiversity, protecting unique nature biotopes, reserving large areas for nature parks, etc., have also been given special thought.
Forest management is a dynamic and long-term process along which the views of citizens change. Today, the ecological treatment of forests is of paramount importance. The maximum wood production, the use of the most efficient wood harvesting methods as well as securing employment to the people living in the country-side, are no longer the most essential and single goals of forest management.
Furthermore, the environmental issues of forest management are also key topics of an active forest research. Scientists need, for example, to find quantitative values for the various forms of forest utilization, identify the ecological consequences of alternative silvicultural actions, develop the thinning and regeneration regimes to be more ecological, improve the tending of seedling stands and monitor the negative effect of acid rain on forests. One interesting topic for research and development activities has been the development of environmentally sound technology for forestry. In this context, Finns are proud to present an approach that could be successfully adopted and used in any part of the world.
FOREST MANAGEMENT PLANNING
In the management of Finland's forests, the environmental protection aspects have always been duly considered by forestry staff. A sustained yield from the forests has never been possible without considering the requirements set by nature. Sixty-two percent of Finland's land area is classified as forests, and 87% of the total area is the so-called "forestry land" serving as drainage areas for lakes and rivers. Thus, the benefit from natural resources, especially from forests, swamps and waters, essentially depends on the activities led by forestry professionals. The meaningful utilization and protection of these renewable natural resources concern, one way or the other, forest owners, professional foresters and forest industry.
The following three topics have been mentioned (e.g., IUCN, WWF, UNEP) as the leading principles for the utilization of natural resources.
_ Protection of basic living functions and ecosystems of the nature;
_ Sustained utilization of ecosystems and the living organisms of them;
_ Securing of the preservation of the manifold and genetic inheritance of living organisms.
From the forest ecological point of view, all actions which support the natural development of forests should be favoured. A joint use of forests for wood production and recreational purposes that simultaneously maintain the prevailing landscape is the goal to be pursued.
Since one third of Finland's land area is covered with peat, the drainage of peatlands, as well as the fertilization of both mineral and peat soils for promoting tree growth, have been common topics in public discussion among the many environmental issues.
Obviously, environmental protection and silvicultural measures are always going to be somewhat in conflict insofar as many environmentalists consider that only reserved parks are natural forests. However, a type of forest management that relies on native tree species and follows the common and quite well-known laws of nature, can successfully fulfil the target of environmentally sound forest management. The speed at which stands and single trees develop can be favourably influenced by human interventions. And today, changes in the landscape take place mostly due to silvicultural measures rather than natural phenomena, e.g., fires, like in passed centuries.
The meaning of clear-cutting operations is not only to harvest the wood crop grown in an area, but also to make the necessary reforestation possible. Trees are unable to live forever and thus young generations are needed to preserve the forest. There are several important planning aspects to be considered, e.g., when drawing the border lines for a cutting area, or when deciding on the size of each uniform reforestation area.
The stand establishment may be carried out either in a natural way, by help of seed trees, or artificially, by sowing or planting. Here, the well-chosen soil preparation method, leaving less-useful trees standing for birds in a clear-cut area, etc., may soften the shock caused by the emergence of a reforestation area in the landscape. Later, the treatment of the seedling stand, if at the right time, can favour the closing of the canopy of trees.
Thinning operations have always been carried out according to efficient forest management. For example, in a mixed stand, selective cutting regulates tree species composition and favours preferred species. On the other hand, if some minor spots are left untouched, good areas for game life are provided, thus breaking _ at least partly _ a standardized treatment of the stand. For the thinning operations, work methods need to be chosen carefully so as to avoid the construction of broad strip roads, damages to the remaining trees, etc.
There always are some points in the landscape which, from a scenic point of view need to be kept untouched or handled with special care. In Finland, the islands with thousands of lakes and the coastal forests of the Baltic Sea are good examples of this. Some rocky areas or other uncommon types of terrain are sometimes also worth receiving attention when deciding on the stand treatment in forest management (Hänninen et al. 1992).
In an efficient forest management, some basic improvement work will also be done. In Finland, operations such as drainage of waterlogged sites, fertilization and construction of forest roads form part of such work. Avoiding erosion and leakage of nutrients to water courses as well as locating roads in the forest area according to the features of terrain, are aspects that need attention (HÄNNINEN et al. 1995).
It is worth stressing that environmental protection and forest management are not separate issues. When planning forest management, the planner must consider the important environmental protection aspects and work out the plan in such a way that there are no conflicts. In most cases, this is fully possible, e.g., in state forests (Korhonen et al. 1994).
DEVELOPMENT OF WOOD HARVESTING
The mechanization of forest work has been rapid during the past decades. More than 80% of the manual wood cutting has been replaced by harvester operations. The transport of wood by horses was discontinued some 20 years ago. The agricultural tractors and later the specially designed forest tractors, the forwarders, have been in use for the off-road transport of wood. The main reasons behind this development have been the higher productivity and the lower costs resulting from the use of these new methods. However, eventual damages caused to nature and other environmental aspects are now gaining increasing importance in the decision-making process when determining what actions to take and what machinery to use in forest operations.
When the manual wood cutting with chain saws was practised, no serious complaints regarding environmental protection were expressed by public opinion. Ergonomics of the forest workers have improved with the use of better tools and better working methods. Today, the introduction of vegetative instead of mineral oils for chain saws is under discussion and in wide use. There are technically acceptable vegetative oils on the market, but their price is still somewhat a problem.
As soon as mechanized wood cutting methods were introduced, damages caused to the remaining trees and to the soil of the cutting site were recognized. In the case of thinning operations, for example, the machinery used in old times was clumsy and the operators, used to clear-cutting operations, were not accustomed to silviculturally selective cutting and they often damaged the remaining trees. With the introduction of cranes and forwarders, machines' movement could be restricted to strip roads thus reducing damages on standing trees considerably, as well as limiting soil compaction. Recently, with improved mobility, larger and broader tires, wheeled tractors are able to manoeuvre, even on a quite soft unfrozen soil, without causing excessive rutting and soil compaction.
REQUIREMENTS SET TO THE RESULT OF WOOD HARVESTING IN THE ENVIRONMENT
Under this section, attention will only be paid to the effects wood cutting has on the environment and to aspects related to the remaining stand; the strip road network, and the eventual damages being caused to the trees and the ground.
The trees left in a selective cutting must be able to fully utilize the wood production capacity of the site. The density of the stand should not be too low after the cutting. The quality of the trees should genetically represent the best part of the unthinned stand. There are recommendations for stand density based on the height of the trees and the basal area.
The strip road network should be designed in such a way that all timber being cut can be transported away from the forest in due course. The spacing of the strip roads should be as wide as the reach of the crane and the forwarder allows. In Finland, the spacing is 20 to 30 m depending on the harvesting method. The width of the strip road should also correspond to the harvester and transport vehicle being used. In young thinning stands, the width of strip roads is 4 meters at the most.
A system and a code have been developed for measuring the damages on the remaining trees (Lilleberg et al. 1991). These are classified as stem, root collar and root wounds. Only a wound located closer than 70 cm from the stem and the wounds in roots thicker than 2 cm are recorded. Wounds of about 100 cm2 of area and bigger ones are recorded separately. The depth and length of the ruts caused by the vehicles are measured systematically.
For the use by the authorities, the code includes two steps:
_ a general evaluation of the results in the cutting area; and
_ a detailed measuring system, if considered necessary, on the basis of the evaluation already done.
The evaluation includes a visit to the field with a logging map for orienteering. The ocular observations are completed with relascope measurements and tape. The report from this survey is delivered to the forest owner, the logging body (e.g., the contractor) and the association of the forest owners.
The detailed measuring includes a systematic survey of the whole cutting area with accurate recording of the basal area and other stand data as well as strip road characteristics, and eventual damages. The results of the survey are then compared with guiding limit values that indicate whether the operation has been carried out successfully and in an acceptable manner. Further actions naturally depend on this comparison. In general, it is sufficient to evaluate a small proportion of the cutting site following this method.
WASTE FROM MECHANIZED OPERATIONS
In mechanized wood harvesting operations, waste liquids, scrap metal and other materials such as rubber are produced from time to time. Some of them are classified as problems and need special attention.
Many logging machines include a great amount of hydraulics; thus there is always the risk to have oil leakages when a pipe breaks or when other damages in the system arise. Even during service operations, careless handling of fuel or oil may lead to spilling these liquids into nature. A system whereby used mineral oils are disposed of in a safe and acceptable manner should be found, particularly as developments in the quality of lubrication are quite interesting. Vegetative oils for example, produced from turnip, rape, or pine oil, are not poisonous and dissolve biologically.
Moreover, when broken machine components are replaced, the scrap material should not be left in the forest nor at sides of forest roads. Too often empty barrels, broken tires, left-overs from an abandoned vehicle, etc., are found when travelling in the countryside. In Finland, a public system exists to clean roadsides, charging the costs to those who littered the streets and highways. Recycling is extensively winning ground today.
CONCLUSIONS
As indicated above, there are numerous and complicated problems when looking at the environmental aspects of forest work. It is obvious that mistakes have been made in the past. However, foresters are in the advantageous position to be able to learn from mistakes and, accordingly, improve future operations. The author of this paper is convinced of the progressive developments being achieved in these areas. Moreover, by utilizing results in the right way, a balanced integration of well-managed forestry and nature protection can be guaranteed. A sign of such development is the recent introduction of a quality control system, according to the ISO 9002 standards, to wood harvesting operations in Finland.
REFERENCES
Hänninen, E. et al. 1992. Metsäammattilainen. Ympäristönhoito (Forestry staff. Environment protection). Metsäteho. Helsinki. ISBN 951-673-128-7.
Hänninen, E. et al. 1995. Metsätalous ja vesiensuojelu (Forestry and protection of waters). Metsäteho. Helsinki. ISBN 951-673-142-2.
Korhonen, K.-M. et al. 1994. Forestry environment guide. Finnish Forest and Park Service. Helsinki. ISBN 951-45-6809-5.
Lilleberg, R. et al. 1991. Harvennushakkuiden korjuujäljelle asetettavat vaatimukset sekä korjuujäljen arviointi ja mittaaminen (Requirements set to the result in thinning and the evaluation and measuring of the result). Metsäteho. Helsinki. ISBN 951-673-119-8.
by
David Quirós, José Joaquín Campos A., Fernando Carrera,
Froylán Castañeda, Robin aus der Beek
CATIE/COSUDE, Turrialba, Costa Rica
Abstract
This paper discusses CATIE's experiences in the development of forest harvesting operations that aim at reducing the impact on the forest. The main factors affecting harvesting operations in the region (e.g., structure and composition of the forest site, and cultural, economic, institutional and political factors) are discussed. It shows how traditional forest harvesting operations in the region are in general highly destructive, mainly due to lack of proper planning and supervision in the implementation of the different operations, and lack of a competent and properly motivated workforce.
It discusses the different harvesting practices that are being implemented by CATIE throughout the region. These practices are grouped into three phases: pre-harvesting, harvesting, and post-harvesting. The pre-harvesting phase includes the following activities: preliminary inventory of the management unit; preparation of the general forest management plan (including the segregation of protection and production areas); planning inventory for the harvesting unit; preparation of the harvesting plan and workforce training. The harvesting plan includes, among other activities: selection and location of all commercial trees, seed trees, trees of endangered species, trees that are recognized as important for wildlife, as well as reserve trees for future harvests; forest roads design and elimination of vines before felling. The harvesting phase includes: construction of forest roads; directional felling; skidding; bucking, loading and transport; and the proper supervision of the implementation of these activities. Finally, the paper discusses the post-harvesting phase, which includes the recovery of timber residues; road repair and closure; and general site cleaning activities (removal of oil cans, fuel containers and other man-made debris).
KEYWORDS: forest harvesting, forest management, tropical forests, Central America.
INTRODUCTION
Since the 1992 United Nations Conference on Environment and Development (UNCED 1992), there have been several initiatives aimed at improving the management, conservation and sustainable development of all types of forests (CIFOR 1995). In Central America, where almost two thirds of all forests have been lost (Pedroni and Flores 1992), particularly the humid tropical forests, there is a growing recognition that managing this kind of forest has become a more complex and more demanding discipline. As a consequence, the design and implementation of environmentally sound forest operations must consider the multiresource character of these forests.
Four elements are essential in relation to low impact forest harvesting systems: comprehensive harvest planning; effective implementation and control of harvesting operations; thorough post-harvest assessment; and development of a competent workforce (Dykstra and Heinrich, 1995).
Since 1990, the Tropical Agricultural Research and Higher Education Center (CATIE)/, has been designing and implementing applied research on different aspects related to the sustainable management of the tropical forests of Central America. The research conducted by CATIE has been carried out with the objective of contributing to the development and adoption of forest management systems that are ecologically sustainable, technically feasible, and economically attractive. Besides these criteria, forest management recommendations should be socially acceptable. The objective of this paper is to describe CATIE's experiences in the development of environmentally sound and cost-effective forest harvesting practices for the broadleaf humid forests of the region.
HARVESTING OF BROADLEAF HUMID FORESTS IN CENTRAL AMERICA
In order to better understand the problematic characteristics involved in logging operations in the region, the main factors that influence timber harvesting will be briefly analyzed.
Factors Influencing Timber Harvesting
Forest biomass
The humid broadleaf primary forests of the region are characterized by a negative exponential distribution (inverse "J" curve) with respect to the number of trees by diameter class. Basically, this distribution describes a forest with few large trees per hectare and an abundance of understorey vegetation and natural regeneration.
On average, these forests contain about 400 trees/ha over 10 cm dbh, and between 30 and 40 individuals greater than 50 cm dbh, which is generally the minimum diameter cut in the region. From those trees greater than the minimum diameter cut, 2 to 10 commercial trees are normally extracted in timber harvests.
These Central American forests are rich in tree species, and generally contain between 100 and 150 different tree species per hectare, although the number of species actually harvested is far fewer. Nevertheless, with recent wood scarcities in many countries, non-traditional species are being increasingly logged.
Geophysical factors
The annual precipitation averages at least 3,000 mm in the majority of Central American humid forests, with the heaviest rains falling between May and December. This long rainy season limits logging operations to only the driest months of the year.
The soils in areas still forested are, in general, infertile with high clay content, thus making them very susceptible to compaction, especially when wet. This factor restricts the machinery that can be used.
The combination of the aforementioned factors makes logging in tropical forests much more difficult than in temperate areas, before taking into account topographical differences. The remaining forests available for harvesting outside national parks are usually located in steep inaccessible areas, with no or poorly maintained road networks. A lack of adequate technology throughout the region further complicates logging activities. The sum of these factors along with the ever increasing agricultural frontier has left only small patches of unlogged forests.
Forest roads are generally constructed by loggers to gain access to valuable trees and as a consequence, rural people take advantage of these access points in their search for new agricultural land.
Socio-economic factors
An agricultural and cattle culture predominates throughout a large part of the population, with almost no forestry tradition. As a consequence, forests have been viewed not as a renewable resource, but rather, as an impediment to development.
The availability of skilled labour is also a limiting factor to low impact logging. Many tree fellers and heavy equipment operators do not receive training, and the skills they do have are acquired on the job.
With a sharply increasing population driving up wood demand, total forest area has significantly decreased. While this situation has lead to the logging of areas with difficult access, consequently complicating timber harvesting, it has also resulted in an increase in both sawn wood prices and the utilization of new species, along with the promotion of management in secondary forests.
Legal and institutional factors
In the majority of harvests, the logger is not the property owner and, as such, has no financial incentive to protect the remaining trees in the stand. Many forest landowners view logging as a means to cheaply obtain access roads on their property. They sell their harvestable trees at very low prices, with the idea of eventually changing the land use to more intensive agricultural or cattle applications.
Land tenure is one of the principal problems involved in sustainably managing Central American forests. The national forest services in these countries will not authorize logging permits unless the property owner can show legal right to the land. Many rural people cannot afford the fees required to obtain government land titles, thus encouraging illegal logging on their lands.
Government policies in general have deterred forest management. Less than a decade ago, property owners requesting bank loans were required to show that they had "improved" their lands, usually by deforesting and planting pasture grasses. In some Central American countries, all standing timber trees were considered state property, and government officials decided when and how many trees to sell on private land. However, now some Central American countries are promoting financial incentives to land holders provided that they put their forest under sustained management. While this new attitude is encouraging, many governments lack the funding to successfully implement these programmes.
Another major impediment to sound forest management practices is the lengthy, confusing, and costly bureaucratic paperwork required by most forest services in order to grant logging permits. Because of this problem, a great deal of the wood sold is illegal, consequently, exacerbating government corruption.
Characteristics of Traditional Timber Harvesting in the Region
Destructive character
The lack of planning and control in timber harvests, in the majority of Central American countries, leads to an activity that is destructive in character, working against principles of sustainability. Generally, the logger assumes all of the administrative responsibilities involved in timber extraction, under the condition that the standing trees are sold at a very low price, greatly under the real market value.
In some countries, this situation is slowly changing with the creation of new laws that require foresters to plan and better supervise logging operations. At first, many loggers resisted these changes; however, some are realizing that a better planned operation actually favours them financially.
Species selectiveness
In the region, timber extraction is characterized by its highly selective nature (high grading), and the number of species harvested is inversely proportional to the area still left forested. For example, in Costa Rica where only 4.3% of the total surface area is still forested (excluding protected areas), many species are harvested, as is the case in La Tirimbina, in Northern Costa Rica, where 36 different species are considered commercial species (Manta 1988). By contrast, in Nicaragua, with 18.2% forest cover, many species are not yet considered commercial, as is the case with "gavilan" (Pentaclethra macroloba) and "almendro" (Dipteryx panamensis). An extreme example is the vast forests of the Petén, Guatemala, where until recently, only two species were extracted in logging operations: Cedrela odorata and Swietenia macrophylla.
Lack of an integrated forest industry
Another characteristic of harvesting is the lack of proper integration between the logging operations and the timber industry. The forest industry in the majority of cases is not directly linked to logging operations, but rather hires out these services through third parties (informal loggers). For this reason, the industry has had little incentive to concern itself in how timber harvests are carried out, despite the fact that in the long-term, their interests will be adversely affected by a reduction of the forest cover.
Obsolete and under-utilized equipment
Mechanized logging with tractors and skidders is used throughout Central America. However, the majority of the machinery is obsolete and under-utilized, thus elevating operational costs and causing greater environmental damage. Oxen, on the other hand, are now seldom utilized for skidding, only by rural communities involved in forest management and by small independent loggers. While this method may appear antiquated, it is in fact a good example of appropriate technology, reducing cost and decreasing environmental impacts.
Excessive wood volume wastage
The large number of logs left unextracted is another characteristic of traditional timber harvests. Because loggers usually pay taxes on the wood volume extracted and not on standing volume, they only take the largest logs without defects. Large limbs of commercial volume are not utilized nor are logs with small cavities. Loggers are financially motivated to extract only the best logs in the shortest time, and normally leave over 20% of the cut wood volume unextracted.
DEVELOPMENT OF HARVESTING SYSTEMS FOR BROADLEAF TROPICAL FORESTS
In the process of forest management, operations relating to timber harvesting require great care and call for prudent planning and execution. In order for success, annual operating plans must be written to specifically cover each activity to be undertaken.
This chapter describes management operations undertaken by CATIE in its experimental forests that relate exclusively to low impact timber harvesting. Given the complexity of this subject, no reference will be made to multiple-use management, even though this is also a management objective in many of the experimental sites.
Technical and Operational Framework
Low impact harvesting techniques have been implemented by CATIE in various sites, from secondary to highly degraded primary forests. These forests do not serve just for developing and validating silvicultural techniques, but also for extension and demonstration purposes. This network of research and demonstration areas is located in the following Central American countries: Costa Rica, Guatemala, Nicaragua and Panama. Demonstration areas located in Honduras will be added in 1996.
The technical framework that constitutes a series of aspects involved in low-impact harvesting can be broken down in the following three phases: pre-harvesting; harvesting; and post-harvesting. The following activities are included in each of these phases:
Pre-harvesting:
_ preliminary forest inventory of the management unit;
_ general management plan;
_ planning inventory of the harvesting unit;
_ harvesting plan;
_ workforce training.
Harvesting:
_ construction of forest roads;
_ directional tree felling;
_ skidding;
_ bucking, loading and transport operations;
_ control.
Post-harvesting:
_ harvest of residual logs;
_ maintenance operations,
Pre-harvesting activities
Preliminary forest inventory of the management unit. This type of inventory is normally carried out to provide information for developing or revising a general management plan. The overall goal is to quantify and qualify the tree population, specifically the distribution of abundance, basal area, and volume by diameter class and commercial groups on a per hectare basis.
The inventory plots are laid out randomly or systematically (with or without stratification). Sampling intensity is usually calculated so that the sampling error is no greater than 20% of the true mean wood volume. In Costa Rica, with forest patches averaging between 50 to 100 ha, sampling intensity is usually greater than or equal to 5% of the inventory area and the normal plot size is 0.3 ha (100 x 30 m). Normally, trees greater than 30 cm dbh are sampled using this percentage, with subsampling down to 10 cm dbh at a 2% intensity.
In summary, this type of inventory gives an overall idea of the potential for timber management in a given area, and provides information on how and when to carry out management activities in different areas of the sampled forest.
General management plan. The general management plan, which comprises the whole management unit, is not only written from a technical viewpoint, but also with an administrative, political, and social perspective in mind.
The technical aspects will obviously take up the majority of the plan, and focus on the presentation of detailed maps showing the annual cutting compartments, and protection areas where timber harvesting will not be permitted. Also illustrated in the general management plan are wood volumes to be harvested, species and minimum cut diameters, growth estimations, cut cycles, and a calendar of activities.
Management plans in Central America have traditionally been documents that are extremely long, costly, and written not as a guide for the property owner, but rather as a bureaucratic necessity to obtain logging permits (Sabogal et al. 1995).
CATIE, with financial support from the Central American Forest Action Plan (PAF-CA) and the World Wildlife Fund (WWF), organized an initiative to simplify management plans throughout Central America for broadleaf forests. Through a regional workshop and in consultation with different projects and institutions, a guide was developed for management plans (CATIE 1994). This has since been amended and approved by the Costa Rican Forest Service (Mirenem 1994), and is pending approval in the other Central American countries. The guide consists of the following components:
_ executive summary;
_ objectives of the management plan;
_ duration and revision of the management plan;
_ basic information on the property;
_ results of the forest inventory;
_ limitations for forest management and possible solutions;
_ management activities;
_ protection activities;
_ register system with suggested forms;
_ calendar of activities;
_ maps;
_ annexes.
Planning inventory of the harvesting unit. This component, also called commercial census, is essential in order to write the annual operational plan and is a 100% inventory of all harvestable trees in the logging compartment. It was first applied by CATIE in 1990 (Quirós and Finegan 1994) and adopted in Costa Rica in 1992 (Mirenem 1991).
The commercial census is conducted walking throughout the proposed harvesting unit, marking commercial species over the determined minimum cut diameter. In order to find and mark the harvestable trees more efficiently, the logging compartment is divided into strips or small blocks, usually with a width between 50 and 100 m depending on topography.
The execution of this type of inventory is divided into two main steps: indication of commercial volume, and site mapping.
In the first step, for every harvestable tree in the logging compartment, the dbh, commercial height, natural lean direction, and location is indicated on standardized forms. Also noted is the grade of difficulty in extracting each tree, and the presence of precommercial-sized trees and vines around each harvestable tree.
In the second step, detailed maps of the harvesting unit that illustrate topographic and hydrologic features are essential in low-impact harvesting. The information needed to generate these types of maps is recorded at the same time as the planning inventory (commercial census). The base map should show contour lines, harvestable trees, and rivers and creeks. With this map, a well-planned low-impact harvest is possible that would exclude both steep areas and water courses (protection areas), and establish skid trails and logging decks so that erosion is minimized.
The cost of carrying out this sort of inventory in La Tirimbina, Costa Rica, was US$ 27.00/ha (Quirós and Reiche in press). This inventory has become a very useful tool for the success of forest harvesting operations.
The map below displays the distribution of all commercial trees greater than 60 cm dbh, rivers and creeks, and proposed forest roads. Such a map significantly increases the efficiency in locating the trees to be harvested as well as those to be left as reserve.
Harvesting plan. The harvesting plan or annual operational plan proposed by CATIE has been put into a guide that specifically states the harvesting operations that need to be carried out in a given year, including maps and charts. In order that these operational plans truly serve as a guide to activities for the person who is going to implement the harvesting operations, they must be as simple as possible, avoiding technical jargon.
Detailed maps of the areas to be harvested and protected are key elements, clearly showing where logging, road building, and/or other activities are to take place on the property. At this time, decisions are made regarding the selection and location of all commercial trees to be harvested (including recommended direction of fall), seed trees, trees of endangered species, trees that are recognized as being important for wildlife, as well as reserve trees for future harvests. This plan should also state how the logs will be skidded, loaded, and transported.
Another objective of the operational plan is to assure that the recommendations made in the management plan relating to productive sustainability are carried out. Essentially, this means that the annual plan should include directions that are easily understood in the field on how many trees (or volume) can be cut in each compartment and guidelines for reducing operational costs and the impact on the vegetation and wildlife, soil, and rivers and creeks.
Workforce training. Before beginning the logging operations, a simple training plan is drawn up for personnel without adequate experience. The objective is to present rational reasons under field conditions for applying low-impact harvesting techniques, as well as the different directional felling and soil erosion prevention techniques, equipment maintenance and first aid.
Harvesting Activities
Construction of forest roads
Wherever possible, previously built roads should be renovated along with the construction of any new roads. The road system should consider the topography and concentrations of trees to be extracted. Maximum skidding distances with the equipment to be used should also be taken into account in road layout. The road layout should also consider minimum slopes, in order to reduce costs and damage to the forest. It should avoid crossing fragile areas, such as rivers and creeks, and unstable soils. The size and power of the equipment should be compatible with the needs for each particular operation, so that costs and damage to the forest are reduced.
The road system can be built on a temporary or more permanent basis, depending mainly on the potential for future timber harvests, and the long term management objectives. Temporary roads are built where the present harvestable volume will not pay the added costs associated with a permanent road system.
The logging road system that has been proposed comprises the following elements.
Primary forest roads. Permit heavy trucks all-weather access to logging decks. In some cases a gravel layer 30-40 cm thick is laid on these roads. In all cases roadside ditches are built and a full-bench design is used. Cross-drains are distributed according to the slope. A 3% minimum and 20 to 25% maximum slope are recommended to avoid excessive erosion. A 3 m width has been used in these types of road. Because of the extra costs associated with these roads, they are maintained and of a permanent nature. The total cost of these types of permanent roads with 30-40 cm thick gravel layer in Villa Mills, Costa Rica was US$ 10.00/m.
Secondary forest roads. Provide skidders and tractors access to the forest, and are not usually of an all-weather design (limited to the dry season). A full-bench design is also used and cross-drains are built into the steeper sections of these roads to control excessive erosion. The maximum slope recommended has depended on the equipment capability (skidder or tractor) and the characteristics of the terrain. The width of these roads has been 2.5 m. The total cost of construction of these types of road in La Tirimbina, Costa Rica was US$ 0.50/m (Quirós and Reiche in press).
Skid trails. Are designed in such a way as to minimize the skidding distance to the secondary roads or logging decks and usually run parallel with contour lines. The distribution of the skid trails should take into account the length of the cable used with the winch and also avoid the tractor or skidder leaving the skid trail. A 2.5 m width has been used. Simple roadside channels are built in order to reduce run-off. The use of these trails is restricted to the dry season and road cuts and fills are avoided.
Given the fact that road construction is the primary cause of erosion, and the main culprit in damaging the residual stand, the overall goal should be to try and maximize the area covered with the least amount of road distance.
In Central America, forest road construction has received little attention, partly due to the propensity for maximizing profits in the short-term.
Directional tree felling
Utilizing the base map with the trees to be harvested clearly marked, the tree fellers do not have to spend valuable time searching for these trees. This is especially important where the harvestable trees are isolated from each other (eg. Petén, Guatemala). Directional tree felling encompasses a series of preliminary steps:
_ cleaning of underbrush around the base of the tree, including neighbouring vines;
_ revision of the natural lean of the trunk and determination of the most adequate felling direction;
_ determination of escape route and brush cleaning if necessary;
_ the actual execution of the cut, employing wedges where necessary; and
_ trunk cleaning and bucking/crosscutting.
The objectives of directional tree felling are based on the following criteria.
Protection of the residual stand. The main emphasis is to minimize damage to the surrounding vegetation, particularly the remaining commercially valuable trees. Also, large tree felling gaps and trees falling in water sources should be avoided.
Ease of log extraction. Where possible, trees should be felled to facilitate skidding, positioning the logs at 30º to 60º angles to the skids trails. This process will also minimize damage to the remaining trees and vegetation in general.
Product protection. This entails the use of special cuts in order to avoid trees falling in difficult terrain conditions or on any kind of impediments that could result in wood splitting and cracking, or the improbability of extraction.
Operator security. Care is taken to minimize risks and tree fellers should be informed of the whereabouts of all personnel and designated escape routes.
In some cases, wedges are needed to correct the direction of the tree fall. In an experimental forest dominated by high volume oaks (Quercus copeyensis and Quercus costaricensis), CATIE is testing five different directional tree felling methods. One of these methods uses cables and manual winches to control the direction of tree felling. However, this method has been necessary in only 1% of the cases.
Before the felling operation takes place, and preferably at least six months in advance, the vines attached to the harvestable tree are cut, usually with a machete.
Measurements of total damage (recoverable and irrecoverable) to the residual forest expressed in terms of percentage of the basal area, in the experimental oak forest of Villa Mills and the experimental lowland forest of La Tirimbina, were 5.9% and 4.9%, respectively (Beek et al. 1992). This damage is much lower than the damage resulting from traditional harvesting operations in the region (usually 30-40%).
Skidding
Mechanized skidding comprises two phases: 1) from the stump to the skid trail, and 2) from the skid trail to the log deck. This first step has been performed with bulldozers using a 30 to 50 m cable with winch. The length of the cable has been increased up to 70 m using a cable extension.
This gives the following advantages:
_ the disturbed area is minimized, thus reducing erosion and decreasing the cost that would be needed to construct skid trails right up to the logs;
_ areas otherwise inaccessible along the length of the winch cable can be harvested;
_ there is more freedom in selecting the optimum route for log extraction; and,
_ the ruts created by the winched-out logs recover more quickly than skid trails.
In the second phase of the skidding operation, agricultural tractors or ideally skidders should be used and not bulldozers, as are often used in the region. Bulldozers are not recommended because of their slowness, thus increasing turn-around time. Also, because of their weight and lack of manoeuvrability, these machines tend to have a more negative impact on the site than lighter, more manoeuvrable skidders or tractors.
In small-scale operations and where there is low volume per hectare, the lower cost option of skidding with animals (oxen, horses, or water buffalo) should be considered.
Bucking, loading, and transport operations
On the log deck, the skidded logs are cut into lengths acceptable by the industry and also according to their transportability. To facilitate loading, a ramp is usually constructed so that logs can easily be pushed onto the truck bed. Mechanical log loaders are rare in the region.
In order to maximize wood yield in terms of acceptance in the market, the logs have been sometimes bucked starting at the small diameter end, measuring and bucking toward the larger diameter.
Control
To assure that the recommendations and established guidelines made in the management and operational plans are followed, a registered forester ("regent forester") should be on site during the majority of the tree harvesting operation. Decisions about modifications needed to the operational plan are made by this person.
Post-Harvesting Activities
Harvest of residual logs
After timber harvest, many logs are left unextracted due to difficult access or because they are split, twisted or are small-size logs. This residual wood can present additional yields for a small-scale operation. In many areas, these unextracted logs are sawn on-site with chainsaws and saw guides, with the products being used for home-consumption or sold on the local market. Quirós and Finegan (1994) estimated for a harvesting operation in La Tirimbina, Costa Rica, that a volume equivalent to 20% of the extracted wood volume came from these kinds of residual logs that were sawn on-site. Other measurements in Corinto, Costa Rica, showed that this figure could be as high as 25% of the extracted wood volume.
Maintenance operations
In the final phase of low-impact timber harvesting, the forest roads that will not be used again in the near future are closed off to avoid erosion, whilst those that will be used in the near future for management or protection of the forest are kept in good condition, paying special attention to drainage. Fences are repaired, tree crowns are removed from any streams or rivers, and any inorganic materials, such as oil cans, fuel containers and other man-made debris left at the harvest site are also removed. These activities are carried out either during or after the harvesting operations.
CONCLUSIONS AND RECOMMENDATIONS
Managing sustainably natural forests in Central America has become a more complex and more demanding discipline, making it essential to design and implement environmentally sound and cost-effective forest harvesting operations. These operations are the first and most important step towards sustainable natural forest management in the region.
In Central America, particularly on private lands, cost-effective and environmentally sound harvesting operations are crucial for reducing the conversion of forest lands to other land uses. By reducing the rate of land use change, the forestry sector is making a great contribution towards the conservation of biodiversity in the region.
The elements that are essential for low-impact forest harvesting systems are: comprehensive harvest planning; effective implementation and supervision of harvesting operations; careful post-harvest assessment; and development of a competent workforce.
The development of low-impact forest harvesting operations is very recent in the region. In this respect, the process that CATIE is promoting for the design and implementation of the guide for management plans, constitutes an excellent means for the transference of low-impact harvesting technology.
The type of environmentally sound harvesting operations described in this paper are being gradually adopted by forest owners and loggers, since their implementation has not resulted in higher costs for forest management operations and has improved the condition of the remaining forest.
In Central America, forest road construction has received little attention, partly due to the propensity of the logger to maximize profits in the short-term and to the lack of adequate knowledge among professional foresters. More attention should be given to this aspect.
Better criteria are needed for the selection of reserve trees, including ecological and production aspects.
The limited development of the forest industry sector in the region in terms of the small number of species processed, restrictions in log dimensions and limited development of timber uses, has become an important disincentive for the development of better harvesting operations. Low timber prices and high transport costs also have a negative effect on the sustainability of the management of natural forests in Central America.
The development of effective forest harvesting systems in Central America must consider the multiresource character of the broadleaf natural forests. In this respect, the harvesting techniques of non-timber forest products (NTFP) that are being developed by CATIE should be included in a code of forest harvesting practice for the region.
REFERENCES
Beek, R. aus der., Quiros, D., y Stadtmüller, T., 1992: Principios, experiencias y resultados de aprovechamiento forestal controlado en dos tipos de bosque tropical con énfasis en la reducción de daños al rodal remanente. Memorias del Foro Internacional sobre los Aprovechamientos Forestales en Selvas y su Relación con el Ambiente. Chetumal, Quintana Roo, México. pp. 2-5.
CATIE, 1994: Modelo de simplificación de planes de manejo para bosques naturales latifoliados en la región de Centroamérica. Turrialba, Costa Rica. 29 p. + anexes.
CIFOR, 1995: A way forward to sustainable development. Forest Research. Jakarta, Indonesia. 28 p.
Dykstra, D. P. and R. Heirnrich, 1995: FAO model code of forest harvesting practice. FAO, Rome. Forest Harvesting and Transport Branch. Forestry Department. 105 p.
Manta, M. I., 1988: Análisis silvicultural de dos tipos de bosque húmedo de bajura en la Vertiente Atlántica de Costa Rica. Tesis Mag. Sc. CATIE, Turrialba, Costa Rica. 150 p.
Ministerio de Recursos Naturales, Energía y Minas., 1991: Guía para la elaboración de planes de manejo y estudios de capacidad de uso de la tierra. Dirección General Forestal. San José, Costa Rica. 25 p.
Ministerio de Recursos Naturales, Energía y Minas., 1994: Guía para la elaboración de planes de manejo y estudios de capacidad de uso de la tierra. Dirección General Forestal. San José, Costa Rica. 98 p.
Pedroni, L., y Flores, R., 1992: Diagnóstico forestal regional para Centro América y propuestas de trabajo. Informe de Intercooperation y UICN/ORCA. San José, Costa Rica. 92 p. + anexes.
Quirós, D., y Finegan, B., 1994: El manejo sustentable de un bosque natural tropical en Costa Rica: definición de un plan operacional y resultados de su operación. Serie Técnica. Informe Técnico no.225. Colección Silvicultura y Manejo de Bosques Naturales no. 9. CATIE, Turrialba, Costa Rica. 30 p.
Quirós, D., y Reiche, C., Análisis financiero de un modelo de manejo sustentable para un bosque natural tropical en Costa Rica. Serie Técnica. Informe Técnico no.__. Colección Silvicultura y Manejo de Bosques Naturales no.__. CATIE, Turrialba, Costa Rica. (in press)
Sabogal, C., Martins, P., y Flores, J., 1995: Planes simplificados de manejo forestal: una propuesta para los bosques latifoliados de América Central. Revista Forestal Centroamericana (Costa Rica), 10, pp 27-32.
by
C.J. Inglis, G. Sutton and G.J. Lawson
Edinburgh Centre for Tropical Forests
Edinburgh, Scotland
INTRODUCTION
The Barama Company Limited (BCL), a joint venture between South Korean and Malaysian investors, signed a contract in 1991 with the Government of Guyana under which BCL was granted a timber concession of 1.7 million ha. The concession was granted for 25 years with an option to extend for a further 25 years. BCL has built a plywood mill outside Georgetown, which currently has 4 lines operational, but plans to construct a total of 12 lines , making it the largest plywood mill in South America. It has been the policy of BCL from the outset to manage its forest concession sustainably. The company recognised the need to have a research and monitoring programme carried out independently of its operational activities, to ensure that all possible precautions are taken to safeguard the environment from avoidable damage.
BCL retained the services of the Edinburgh Centre for Tropical Forests (ECTF) in 1992 to undertake a programme of monitoring and research. The primary objective of the project is to carry out independent monitoring of the activities of BCL against a series of environmental criteria and to conduct a research programme which will inform the forest management plan and assist the company to fulfil its objective of: "ensuring the sustainability of the forest in perpetuity whilst selectively harvesting a major natural resource for the benefit of the company, people and Government of Guyana".
It is rather early to provide firm conclusions on the sustainability of the company's activities, but this paper describes the impact assessments, monitoring, research and training activities which are being undertaken by ECTF. Most emphasis has so-far been placed on monitoring the effects of BCL's current logging practices on tree cover and composition. Research plots on modified logging techniques have been established, but development of this part of the programme is anticipated for the future.
ENVIRONMENTAL BACKGROUND
The BCL concession is located in the north-west of Guyana (Figure 1), in an area which is remote from centres of population and can only be accessed by air or river.
The Physical Environment
The concession is almost entirely underlain by the pre-Cambrian plateau of the Guiana Shield. The major geological formations in the concession are granite and greenstone. The soils in the northern part of the concession are acid clays of low fertility. Three major rivers dissect the area: the Barima and Barama draining to the north in Venezuela, and the Cuyuni draining centrally into the Essequibo River. All rivers and streams drain into these, except for small watersheds draining into the Kaituma River on the northern concession boundary, and into the Puruni River on the southern concession boundary.
The concession is predominantly flat or undulating lowland without major topographic features except in the west. On the eastern fringes of the concession there are some patches of coastal swamp and towards the south-east there are patches of white sand. Most of the concession is dry land with intermixed marsh forest and riverine flood plain. The western parts are more hilly but nowhere does the topography exceed 240 m.
The climate of the area is fairly typical for lowland rain forest, with an annual rainfall in the northern part of the concession of 2,400 mm. It rains during every month with rainfall peaks in June-July and December-January, during these periods average monthly rainfall exceeds 300 mm. The driest period is March-April with average monthly rainfall of 90 mm. Mean temperatures range between 23 and 32_C and relative humidity is between 83 and 97%.
Figure 1. Location of the BCL concession in northwest Guyana
Guyana is 75% forested (over 16 m ha) of which 3.6 m ha is currently accessible for timber production. Timber volumes are typically around 180-220 m3 ha-1 (all species), with 100-150 m3 ha-1 of "commercial" species (FAO 1970). Greenheart (Chlorocardium rodiei) is the dominant commercial species, comprising 1.5% of total timber volume, 40% of timber harvested, and 70% of timber exported. Mean stocking density for the mixed forest is 25 commercial trees ha-1 out of a total stock of 75-100 trees ha-1 (>20 cm dbh).
The concession area is almost entirely forested apart from modest areas which have been cleared for mining and past agricultural development. No forest inventories have previously been undertaken in the concession area, nor has any research taken place on the impacts of logging.
Greenheart is largely absent from the BCL concession, as are several other timber species from central Guyana like Wallaba (Eperua sp.), Morabukea (Mora gongrijpii) and Wamara (Swartzia leiocalycinai). Forest types in NW District are dominated by "mixed dryland forest" (Table 1), although the BCL inventory suggests a higher proportion of seasonally flooded (20%) than previously thought.
Table 1: Proportions of forest types in the North-West Region (Welch 1975)
Mixed forest on undulating or hilly terrain Mixed forest on deeply dissected terrain Mora forest on flat, seasonally flooded riverine terrain Mixed forest on steep hills Wallaba forest on white sands ridges Mixed forest on flat terrain along rivers |
79% 13% 4% 3% 0.5% 0.5% |
BCL has undertaken an inventory of merchantable timber in the NW part of the concession, this was initially a 100% sample (Table 2), but is now 20%. The initial inventory counted only undeformed stems of commercial species with diameters over 35 cm. This showed the commercial volume to be around 34 m3 ha-1, with 70% in plywood or potential plywood species. The size frequency distribution (Figure 2) is heavily skewed towards the smaller age classes. Baromalli and Crabwood occur in highest densities in dry forest, and white Cedar (Tabebuia insignis) and Mani (Symphonia globulifera) in marsh forest. Few trees have diameters exceeding 100 cm, and they have a lower stature than in other parts of Guyana.
Guyana has 72 species of endemic higher plant species (GAHEF 1992), and 10 endemic amphibians and reptiles. 144 Guyanian animals and plants are listed in CITES (Convention on International Trade in Endangered Species) No measurements of biodiversity have been made in or near the concession, although an ethnobotanical survey has identified plants of use to man and certain animal groups (Adams 1972).
Figure2. Frequency distribution of commercial trees in 3 x 100 blocks typical of the NW section f the Barma concession
Table 2: Commercial species by volume (100% sample of 4,510 ha)
Common name |
Commercial use |
Botanical name |
Trees/ha |
Volume (m3ha-1) |
% by volume |
Baromalli Haiariballi Black kakaralli Crabwood Kurokai Maho Kabukali Burada Dalli Waiki Wina kakaralli Syradan Bulletwood Maporokon Simarapa White kalaralli Jack-in-the-box Soapwood Futui Monkey pot - |
plywood plywood non-comm plywood plywood plywood timber non-comm plywood - non-comm - protected plywood timber non-comm non-comm - plywood timber - |
Catostemma sp. Alexa sp. Eschweilera sp. Carapa guianensis Protium sp. Sterculia sp. Goupia glabra Parinari sp. Virola surinamensis Inga sp. Eschweilera corrugate Hieronyma laxiflora Manilkara bidentata Ingis alba Simaruba amara Eschweilera sp. Hernandia sonora Pithecellobium jupunba Jacaranda copaia Lecythis davisii 75 other species |
4.8 4.48 3.51 1.29 1.2 0.75 0.35 0.24 0.21 0.41 0.21 0.26 0.06 0.21 0.11 0.14 0.11 0.11 0.15 0.15 2.33 |
10 4.26 3.25 1.36 0.96 0.83 0.81 0.5 0.34 0.31 0.3 0.3 0.29 0.24 0.24 0.22 0.2 0.19 0.18 0.18 2.01 |
37.1 15.8 12.1 5 3.5 3 3 1.9 1.3 1.1 1.1 1.1 1.1 0.9 0.9 0.8 0.7 0.7 0.7 7.5 |
TOTAL |
21.08 |
26.97 |
99.3 |
The Socio-Economic Environment
The area has had few major economic activities, apart from a manganese mining operation which closed down almost 30 years ago. Now the most important activity is gold mining. This is expanding in response to an increase in world price for gold. High pressure hoses are used to cut through the 10-15 m clay banks of rivers, turning river-water bright brown for many kilometres, and causing the death of trees in flooded areas around the mine workings. Mercury is also used to consolidate the sluiced gold, with potential consequences for the fish populations. While mining activities are currently localised and small scale, it is very likely that mining will increase considerably when BCL's forest roads improve the access to more remote areas (Harrison et al. 1993).
Birds, reptiles, freshwater fish and occasionally cats are collected by Amerindians, and many are exported in a trade worth US$ 1.5 million annually (Gahef 1992). Little commercial fishing takes place except for the export of aquarium fish.
The population of the concession is very low (~1500), largely composed of Caribs, Arawaks and Wauru. Literacy is reasonably high, and most families have a cash-based living from mining, agriculture, pork-knocking, government service, logging, farming and training. They seem to welcome the arrival of BCL, which has a policy of employing and training local labour, and bringing perceived benefits of roads, new markets and health care.
MONITORING ACTIVITIES
Environmental and Social Impact Assessment (ESIA)
Prior to the commencement of logging operations and the design of the monitoring and research programme, an Environmental and Social Impact Assessment (ESIA) of BCL's forestry developments was carried out by ECTF (Harrison et al. 1993). The aims of this study were to examine the potential negative and positive impacts of BCL's forestry operations and to produce guidelines and recommendations for incorporation by BCL into improved planning and forest management. The justification for the ESIA was four-fold:
_ Identify potentially serious negative impacts of logging activities on environmental and social welfare, which may impede the objectives of sustainable forest management.
_ Inform management of practices that satisfy wider national and international concern and that satisfy criteria being developed by a number of international bodies for wise forest management.
_ Identify those aspects of environmental concern which need to be monitored in detail during the logging programme.
_ Foster better understanding of environmental and social relationships that may result in improved environmental management, socio-economic benefit and longer term economic growth in the region.
Background information was collated, likely impacts identified, and recommendations for mitigation of these impacts made to the Company
Baseline Information on Tree Populations
BCL's inventories collect information only on non-defective trees of commercial species with diameters exceeding 35 cm in diameter. These data give a good indication of commercial potential, but are of much less use in assessing logging sustainability and stand dynamics.
Thus, ECTF initiated in 1993 a programme to establish pre-logging permanent sample plots (PSP), partly to characterise the smaller diameter classes, and partly to enable logging damage to be quantified, and partly to provide a baseline for future measurements of growth and yield. This monitoring will determine whether the estimate of allowable cut, currently being applied by BCL, is sustainable. The recovery period for different forest types will be measured, following logging. A realistic felling cycle will be established. There is no satisfactory method of calculating the number of PSPs required for multiple-objective data capture, and the number established in the concession is largely governed by the availability of human and other resources. The PSP work was started in May 1993, one month in advance of the commencement of commercial logging operations. During the first 12 month period of this programme, 15 PSP were established, and during the second 12 month period a further 17 were established. This represents approximately one PSP per 1,000 ha of forest logged. It will not be feasible to maintain this sampling intensity because of the increasing volume of plot remeasurement work which is required. In the current 12 month period, the work plan includes provision for the establishment of a further 10 PSP which will represent approximately one plot for each 2,000 ha of forest logged during the period. The methodology for plot establishment and measurement is fully described by Alder and Synott (1993), and involves measurements of trees >20 cm dbh and subsamples of poles, saplings and seedlings (Figure 3). Additional m2 sub-quadrats of seeds are monitored in each of the 25 quadrats.
15 |
25 |
35 |
45 |
55 |
14 |
24 |
34 |
44 |
54 |
13 |
23 |
33 |
43 |
53 |
12 |
22 |
32 |
42 |
52 |
11 |
21 |
31 |
41 |
51 |
Figure 3. Measurement protocol for each 100 m x 100 m Permanent Sample Plot (PSP). All trees >20 cm dbh measured in each quadrat, poles (5-20 cm) measured in 23,34,33,32,43; saplings (dbh<5 and ht>1.5 m) measured in 34, and seedlings measured in 33.
The forest in North-West Guyana is characterised by its small stature. An analysis of the PSP data reveals that the average canopy height is approximately 27 m and that stem diameters are relatively small (Table 3). The average diameter for all trees with a diameter greater than 20 cm is approximately 37 cm.
Table 3: Diameter distribution per hectare (average of 34 PSPS)
Diameter class (cm) | ||||||||||
20-29 |
30-39 |
40-49 |
50-59 |
60-69 |
70-79 |
80-89 |
90-99 |
100+ |
Total | |
No. of trees % |
108 50 |
54 25.2 |
29 13.5 |
12 5.7 |
6 2.7 |
4 1.7 |
1 0.6 |
<1 0.3 |
<1 0.3 |
215 100 |
The minimum diameter limit for general harvesting purposes in Guyana is 35 cm and only 32% of the growing stock (70 trees per ha) is represented by trees above that diameter. The ply mill demands logs larger than 35cm, and BCL sets lower felling limits of between 50 and 60 cm, depending on species and bark thickness. There are, on average, only 7 trees per hectare which satisfy the species, quality and size conditions, and BCL is in fact felling only an average of 5 trees per hectare, with an average volume of 3 m3 per tree. Thus this a low yield in comparison to other forest in the Amazon Basin. In North-eastern Brazil, for example, an average of 6 trees are harvested per hectare, yielding 38 m3 of timber (Verissimo et al. 1992).
The basal area per hectare ranges between 16.1 and 26.5 m2 with an average of 22.12 m2 (Table 4).
Table 4: Basal area distribution (m2/ha) (Average of 34 PSPS)
Diameter class (cm) | ||||||||||
20-29 |
30-39 |
40-49 |
50-59 |
60-69 |
70-79 |
80-89 |
90-99 |
100+ |
Total | |
Basal area (m2ha-1) % |
4.85 21.9 |
4.82 21.8 |
4.31 19.5 |
2.79 12.6 |
1.75 7.9 |
1.46 6.6 |
0.77 3.5 |
0.46 2.1 |
0.91 4.1 |
22.12 100 |
Twelve species or species groups, from the 140 so far recorded, represent 73% of total average basal area per hectare (Table 5).
Two species, baromalli and haiariballi represent some 90% of the total volume of timber selectively felled by the company. It is encouraging to note, from the perspective of sustained yield management, that these two species are not only common as trees but are also well represented in the classes of seedlings, saplings and poles (Table 5).
At this stage in the project, insufficient data has been collected to allow reliable estimates of incremental growth or the impact of different intensities of logging on residual stand growth to be calculated. It is, however, possible to examine the effect of logging on the residual surviving trees.
Table 5: Basal areas and frequency of different age classes of the dominant species and species groups in the 12 PSPs established in 1993, together with their relative importance in the BCL harvest
Ranking in PSPs (by ba) |
Local name |
Botanical name |
% Basal area |
Trees (ha) |
Poles (ha) |
Samplings (ha) |
Seedlings (ha) |
Ranking in BCL harvest |
1 2 3 4 5 6 7 8 9 10 11 12 |
Black Kakaralli Haiariballi Kauta Baromalli Trysil Kurokai Waiki Yarula Crabwood Maho Bartaballi Kabukalli |
Eschweilera sagotiana Alexa imperatricis Licania spp Catostemma spp Pentaclethra macroloba Protium decandrum Inga spp Aspidosperma spp Carapa guianensis Sterculia pruriens Ecclinusa spp Goupia glabra |
19 12 10 9 6 4 3 2 2 2 2 2 |
50.2 28.3 24.1 11.7 17.8 17.1 11 1.8 4.1 4.1 2.9 1.8 |
131 78 58 30 36 72 26 5 5 12 3 - |
202 196 146 90 319 71 33 10 179 17 4 |
3,177 563 552 2,146 31 1,750 1,063 21 31 1,229 10 - |
3 2 4 1 - 6 - - 5 7 - 8 |
Logging Damage
20 PSPs were logged by BCL in the normal course of the logging operations, are operators are ordered to treat them no differently than the surrounding forest. Following logging, each PSP is revisited to establish the number and basal area of trees felled or damaged in the process of logging. Nine classes of damage are recognised, which are coded as follows:
0 - No damage.
1 - Light bark damage.
2 - Severe bark damage.
3 - Light crown damage.
4 - Severe crown damage.
5 - Crown snapped off.
6 - Tree pushed or pulled over.
7 - Tree felled.
8 - Tree bent over.
Data from 11 of the PSPs have been analyzed (Table 6), with the most significant form of damage being applied where two or more damage types were apparent.
In terms of logging damage reduction, it may be of some significance that while an average of 6 trees per plot were felled, a further 21 were either pushed or pulled over or were snapped. This points to a need for research to be carried out into the cost and effectiveness of pre-logging climber cutting.
Table 6: Summary of damage on logged over PSP
Code |
Damage |
Mean number of trees per ha |
% |
Range |
0 1 2 3 4 5 6 7 8 |
None Light bark damage Severe bark damage Light crown damage Severe crown damage Crown snapped Pushed over Felled Bent |
183 3 1 9 4 4 17 6 1 |
80 1.3 0.4 3.9 1.7 1.8 7.5 2.4 0.4 |
133-229 0-8 0-4 0-23 0-14 0-9 0-37 0.1-7 0-3 |
Total |
227 |
100 |
- |
Canopy Gaps
At the same time that the post-logging damage assessment is carried out, canopy gap sizes are also measured. Under each canopy gap, a datum point is established and the distance and the bearings to a number of reference points at the "drip line" of the canopy edge are measured. This allows each gap to be plotted and its area calculated. The data collected so far indicate that the average gap created during selective logging extends to some 340 m2. This compares favourably with average gap sizes of 600 m2 recorded in Brazil for a logging operation of similar intensity (Verissimo et al. 1992). The average aggregate area of canopy gaps within the PSP, which have been assessed so far, is 950 m2 or 9.5% of total PSP area. This average has been derived from a range of 2,849 m2 (28.5%) and 210 m2 (2.1%).
The data on canopy gaps and logging damage to trees is illustrated using data from the most severely logged PSP from the 11 (Figure 4). Three of the PSPs were not logged at all.
RESEARCH
The PSP programme monitors the response of stands to the current levels of extraction decided by BCL. Initially the average cut was planned 25 m3 ha-1, on the assumption that the forest might grow at around 1 m3 ha-1yr-1 on a 25 year cycle. However, they have remained highly selective in the species, size and quality of log removed, and are currently cutting only 15 m3 ha-1. This represents only 60% of the planned harvest. Clearly, proportion of standing volume extracted is an important variable controlling the sustainability of harvesting. A set of research plots was established to examine the effect of fixed extraction levels on subsequent regeneration growth. The design involves 4 treatments x 9 hectare treatment areas x 3 replicates, with each treatment represented by a central 1 ha plot surrounded by a 100 metre buffer zone. The treatments are: (a) no logging control, (b) reduction in basal area by 10%, (c) basal area reduction by 20%, basal area reduction by 30%.
DISCUSSION
The impact of selective logging on tropical forest is receiving increasing attention in South American forestry, and a particularly important series of reports have been produced by the Centre for Agricultural Research in Suriname, Anton de Kom University of Suriname and Wageningen Agricultural University (de Graff 1986, Jonkers 1987, Hendrison, 1990). These reports are referred to in the BCL Forest Management Plan, and have partly guided present forest operations. However there are a number of opportunities for research plots which have not yet been implemented. In order of priority these are:
_ Trials of climber cutting some months prior to felling: this is likely to significantly reduce damage to adjacent trees. Costs should be collected in parallel with measurements of damage and regrowth.
_ Trials using the winch cable on stumping tractors should assess the extra time and cost involved against the reduction in damage caused to advance regeneration as the tractor manoeuvres repeatedly around the stump to position the butt in convenient positions for the skidder.
_ Trials of improvement thinning at or following harvest - where selected medium diameter trees are favoured by felling or poisoning competing trees of lower value. This may be economic at the time of harvest, but less so after a gap of several years as recommended in the CELOS system (de Graff 1986).
_ Retaining at least one seed tree/ha of each of the two main commercial species in an effort to maintain population profile of seedlings, saplings, and poles necessary for sustained production.
REFERENCES
Alder, D. and Synott, T. J. Permanent Sample Plot Techniques for Mixed Tropical Forest. University of Oxford, 1992.
de Graaf, N. R. A Silvicultural System for Natural Regeneration of Tropical Rain Forest in Suriname. University of Wageningen, 1986.
Hendrison, J. Damage Controlled Logging in Managed Tropical Rain Forest in Suriname. University of Wageningen, 1990.
Jonkers, W. B. J. Vegetation Structure, Logging Damage and Silviculture in a Tropical Rain Forest in Suriname. University of Wageningen, 1987.
Mennega, E. A., Taminens-de Rooij, W. C. M., and Jansen-Jacobs, M. J. (Eds) Check List of Woody Plants of Guyana. The Tropenbos Foundation, 1988.
Verissimo, A., Barreto, P., Mattos, M., Tarifa, R., Uhl, C. Logging Impacts and Prospects for Sustainable Forest Management in an Old Amazonian Frontier: The Case of Paragominas. Forest Ecology and Management 55 (1992) 169-199.
by
Pedro Moura-Costa
Innoprise Corporation Sdn Bhd (ICSB), Sabah, Malaysia, and
Oxford Forestry Institute, Oxford, United Kingdom
Abstract
A commercial project introducing reduced impact logging (RIL) techniques was conducted over 1 400 ha in a forest concession in Sabah, Malaysia. The project was a joint venture between Innoprise Corporation, a Malaysian forestry company, and New England Power, an American electricity company trying to reduce its net CO2 emissions. The objective of the project was to reduce the damage to the residual stand in order to reduce the amount of CO2 emissions from decomposing biomass. Guidelines were developed for logging dipterocarp forests and training activities were carried out. A system for verification of compliance was developed and implemented. Results have shown that the use of reduced impact logging techniques can lead to an overall reduction of approximately 50% in the impact of logging, compared to the traditional methods currently in practice. Initial carbon savings reached 42 tons per ha over two years.
INTRODUCTION
In the process of extracting 8-15 trees (80 m3, _ 22 t carbon) from a hectare of forest in East Malaysia, often as many as 50% of the remaining trees are damaged and up to 40% of the area is crushed by bulldozers (Nicholson 1958, Sabah Forest Department 1989). The effects of uncontrolled logging on biodiversity and ecosystem function are also severe; soil erosion, weed infestations, and incidence of fire all increase in logged-over forests. These effects combine with the destruction of much of the pre-existing regeneration of commercially valuable trees to make uncontrolled logging extremely detrimental to long-term ecological and economic productivity. While the merits of various forest management systems are being debated, tropical silviculturalists are of one voice in advocating the use of reduced impact logging techniques (Palmer and Synnott 1992).
Numerous studies have shown unequivocally that logging damage can be substantially reduced through directional felling and planned extraction of timber on properly constructed and carefully utilised skid trails (e.g., Froehlich et al. 1981, Hendrison 1990). Due to the high cost of operating bulldozers and other ground-based log extraction equipment, controlling the logging process can actually be accomplished at a very low cost (Jonkers and Mattsson-Marn 1986, Hendrison 1990).
Concern about rising atmospheric concentrations of greenhouse gases (GHGs) has prompted the search for methods of sequestering carbon in plant biomass. The signing of the Framework Convention on Climate Change by 153 nations at the 1992 United Nations Conference on Environment and Development (UNCED) (Grubb et al. 1993) is evidence of intent to restrain the build-up of GHGs in the atmosphere. Due to cost effectiveness, high potential rates of carbon uptake, and associated environmental and social benefits, much attention has focused on promoting tropical forestry for offsetting carbon emissions (e.g., Brown et al. 1995).
Damages incurred during conventional timber felling and extraction result in reduced forest biomass and substantial releases of carbon dioxide (Houghton et al. 1987) and possibly other radiatively important gases (e.g., methane). It is possible therefore, to reduce the release of CO2 from logging by adopting less destructive logging techniques. Carbon dioxide sequestration in tropical forests through controlled logging (by reducing emissions) has a number of benefits. Immediate carbon benefits are realised at low expense, native species are maintained on site, and soils are less likely to be degraded or eroded. Post-logging carbon accumulation rates are also expected to be elevated and the future potential for production of timber and non-timber forest products enhanced. These benefits derive from the reduced likelihood of vine infestations and a healthy stand.
This paper describes a project which used reduced impact logging techniques in order to reduce forest damage and CO2 emissions, in Sabah, Malaysia. The background of the project is given, as well as an overview of its technical and economic aspects.
THE PROJECT
Background
The Reduced Impact Logging (RIL) Project is a cooperative venture between Innoprise Corporation Sdn. Bhd. (ICSB), a semi-government organization which has the largest forest concession in the state of Sabah, Malaysia, and the New England Power (NEP) Company, an American utility trying to address the challenge of reducing its net CO2 emissions. The objective of the project was to introduce the use of reduced impact logging (RIL) techniques in order to lower the level of damage caused by selective harvesting operations, reducing the release of CO2 from decomposing vegetation and soil loss (Marsh 1992).
In an initial phase, 1 400 ha of forests were logged according to the RIL techniques from 1992 to 1995. The project aimed at reducing logging damage by 50%, thus saving approximately 36 t C ha-1 at a cost of US$ 2.50 per t CO2 (based on estimate for carbon savings two years after logging, Putz and Pinard 1993). All the incremental costs of training and implementation of the project were paid by NEP, who has full rights to the carbon savings. On the other hand, ICSB benefited from improved management of its forests, and a better residual stand after logging.
The project was located in two compartments within ICSB's logging concession: one near Lahad Datu, and one near Tawau, in Eastern Sabah. Most of the research was conducted at the compartment near Lahad Datu, which is close to the Danum Valley Field Centre (DVFC), a well established ecological research station (Marsh and Greer 1992), which provided logistic and scientific support for the foresters and researchers involved in the project.
The project consisted of four main components. Firstly, development of appropriate logging guidelines for the forest conditions in Sabah. Secondly, training of staff on the various aspects and techniques required for the successful implementation of the guidelines. Thirdly, establishment of a system for independent verification of compliance and quantification of logging damage. Finally, development of a method for quantification of biomass and carbon pool changes after logging. This paper describes these components, as well as some of the results, problems and outcomes derived from the initial phase of the project.
Guidelines for Reduced Impact Logging
Given the inevitable impact of heavy machinery on the forest environment, the key to the success of reduced impact logging techniques is the use of very-well planned guidelines for timber extraction.
The logging guidelines adopted by Innoprise were based on those developed by the Queensland Forest Service, for logging tropical rainforests in Australia (Queensland Forest Service, unpublished). It also incorporated components of the guidelines of the SUAS (Swedish University of Agricultural Sciences) Project (a logging research project jointly conducted with ICSB; Cedergren et al. 1994), the Rainforest Alliance's timber certification program (Smartwood), and the CELOS management system developed in Suriname (de Graaf 1986). The RIL guidelines were customised to address the characteristics of the site: heavily stocked forests (up to 170 m3 ha-1 extractable timber) with very large trees (up to 65 m tall), steep terrain and heavy rainfall (ca. 2 700 mm yr-1).
The guidelines were developed in an interactive and dynamic way, based on a process of discussion, trials, evaluations and continuous improvement. It included ICSB foresters and forest rangers, logging contractors, specialists in logging from the Queensland Forestry Service, SUAS, and University of Florida, representatives of the Sabah Forestry Department, and researchers from DVFC. Given the difficulty of the terrain and logging conditions, the development of these guidelines were a learning process for all concerned.
The guidelines concentrated on reducing the impacts of tree felling and heavy machinery on the remaining timber stock and forest soil. The main elements of the guidelines are as follows:
_ 100% inventory and mapping (1:5000 scale) of trees to be harvested and potential future crop trees (PCTs) >20 cm dbh;
_ Designation and mapping of streamside buffer zones, steep slopes, and sites known to be of importance to wildlife (e.g., large fig trees);
_ Cutting all vines (dbh >2 cm) at least 9 months prior to logging;
_ Marking trees to indicate direction for felling, which should aim at facilitating skidding and avoiding damage to potential crop trees;
_ Planning and marking roads and skid trails, aiming at reducing skidding distances, down-hill skidding, and stream-crossing. Roads and skids were preferentially located on ridge-tops;
_ Reduction in the size and number of log-landings, using road-sides whenever possible;
_ Protection of surface soils and water courses, by reducing use of bulldozer blades, introduction of cross-drainage on skid trails, adoption of a maximum slope limit for tree felling (35_) and skidding (25_), establishment of stream buffer zones, and halting forest operations during wet weather.
The guidelines include detailed specifications for road construction, stream crossing, wet weather shut-down, skid trail width, log-landing size and location, and post-logging closure of roads and skid trails (ICSB records, unpublished document; Putz and Pinard 1993, Pinard and Putz, in press, Pinard et al., in press). None of these elements is original (e.g., Jonkers 1987, Hendrison 1990, Dykstra and Heinrich, 1995), but they have rarely been implemented in tropical forest conditions on a commercial scale.
Training and Implementation
The initial phase of the project included a substantial training component. Training was aimed at all levels in the ICSB hierarchy, including the senior management, foresters, forest rangers, tree fellers and tractor operators. Foresters had to get acquainted with technical and planning tools for improved management of harvesting operations. This included a visit to the Queensland Forest Service, associated consultancy support to assist in the development and implementation of RIL methods, and attendance at specialised courses. Some of ICSB staff are now pursuing further academic qualifications in this subject. Training of forest rangers aimed at a confident understanding of the RIL guideline, in order to ensure its proper implementation on the ground. Rangers were in charge of supervising the logging operations, a task which requires a considerable degree of decision taking. Chainsaw operators were trained on tree marking and directional felling techniques, including a short field course provided by Nordfor, a Swedish training consultancy working with the SUAS project. Bulldozer drivers were instructed on less damaging ways to skid, avoiding use of the blade and maximising use of the winch. On-going continuous training of the logging crews was provided by ICSB foresters and rangers. The major challenge in implementation, however, was one of management, because RIL challenges traditional attitudes and requires a fresh look at operations, procedures, payment scales, and so on.
Verification
Essential elements of the RIL project were the verification of compliance with guidelines and quantification of logging damage. In order to give the project credibility, an Environmental Audit Committee was created including representatives of three recognised institutions in the fields of forestry and environment. These were: the Rainforest Alliance, an internationally recognized non-governmental organization based in the US, which runs the Smart Wood timber certification programme; the Forest Research Institute Malaysia; and Francis Putz of the University of Florida and CIFOR. This committee met at regular intervals during the initial phase of the project in order to evaluate its progress and assist in the refinement of the field practices. Eventually, its final aim was to verify whether ICSB had correctly followed the RIL techniques and if this had led to significant reductions in forest damage.
Carbon Quantification
The primary objective of the project was to reduce the amount of carbon dioxide release caused by the effects of logging operations. In order to determine these figures, a thorough study was conducted by researchers from the University of Florida that quantified the impacts of logging on biomass and carbon pools. This was carried out in eight paired plots, each of which consisted of one logging compartment (ca. 35-50 ha each), which were logged according to RIL and conventional techniques (Pinard and Putz, in press). The central points for determination of carbon pools were: (1) the quantity of carbon retained in standing biomass; (2) the rates at which logging residues decompose; (3) the rates at which regenerating forest sequesters carbon; and (4) changes in soil organic matter content. A simpler method for the rapid estimation of carbon pools in a logged forest was also devised. This method is based on assessments of logging damage and use of a computer model specially designed for this purpose (Pinard 1995).
RESULTS
Logging Damage
The overall conclusion of the Environmental Audit Committee was that the implementation of the guidelines was adequate, and that it created substantial reductions in the logging impacts compared to the conventionally (CNV) logged areas.
Table 1 shows a comparison of the impact of harvesting 120 m3 ha-1 of timber in hilly areas using either RIL or conventional logging techniques. It can be seen that both the length and the width of roads and skid trails were much reduced in RIL areas. Log landings in RIL were about 50% smaller than in CNV logged areas. These figures are even more impressive if compared to other estimates from the region, which found up to 30-40% of logged areas traversed by bulldozers (Sabah Forestry Department 1989, Jusoff 1991, Nussbaum et al., 1995). In RIL areas, roads and skids were better drained and the soil was less disturbed. There was substantially less removal of fertile topsoil since there was much less use of the bulldozer blade. Apart from a smaller number of damaged trees in the RIL areas, the severity of damage was also lower (details given in Pinard and Putz, in press).
The combined effects of RIL resulted in a forest with less disturbed soil, canopy, water courses and residual stand. It is expected that natural recovery in these areas will be much faster than in conventionally logged areas. Some of the skid trails in the RIL area were already covered in vegetation one year after logging while, in some cases, bare skid trails can still be found 20 years after conventional logging in the same region (Pinard 1995).
Table 1: Figures of logging impacts after harvesting of 120 m3 ha-1 in hilly area in Sabah, following different logging techniques. Summary of results from ca. 800 ha
RIL |
CNV | |
Roads |
20 m ha-1 1.6% of logged area |
24 m ha-1 3.3% logged area |
Skid trails |
71 m ha-1 4% of logged area |
205 m ha-1 13% of logged area |
Log landings |
57 m2 ha-1 |
103 m2 ha-1 |
Trees damaged |
29% |
56% |
RIL = reduced impact logging; CNV = conventional logging techniques.
One of the main problems of the project was that the limitations imposed by the guidelines caused a reduction in the overall productivity of the area. The main limiting factor was the slope limit for felling and skidding. Apart from the large number of trees found in steep areas, the slope limitations for skidding in some cases prevented access to areas with lower slopes. This could be circumvented if aerial logging techniques (e.g., Blakeney 1992, Dykstra and Heinrich, 1995) were introduced.
Carbon Quantification
Project results on carbon quantification showed that an average of 42 t C ha-1 was saved in the short term after RIL (Pinard and Putz, in press) which is slightly higher than the original estimates. Further savings are envisaged in the longer term. As the logging debris and dead trees fully decompose, there is going to be a reduction in carbon stored in both areas, but a much larger decrease is expected in conventionally logged areas (where there is much larger amounts of logging debris and damaged trees). As forest regeneration progresses, the amount of carbon stored in both areas will increase. However, higher rates of biomass accumulation are expected in RIL areas, further increasing the differential between both systems. Figure 1 shows a diagrammatic pattern of carbon reduction and re-accumulation after RIL and conventional logging.
Financial Aspects
In an initial analysis, the operational costs of RIL are higher than those of conventional logging, due to the extra activities required, i.e., detailed inventories, training, climber cutting, intensive supervision. However, after the initial phase of learning and intensive training, it is expected that RIL operations will run more smoothly and efficiently than in conventional logging. This may lead to savings through, for instance, reduced use of bulldozers, with lower fuel and maintenance costs. A time-and-motion study of the various activities associated with RIL and conventional logging is being undertaken by ICSB, and will provide a better understanding of the costs associated with these two systems.
Given the project cost of US$ 450 000 and the total savings of 58 000 t C over the project area, the cost of carbon saved would be US$ 7.60 per t C (US$ 2.07 per ton CO2) at two years after logging. Higher savings are expected in the longer term. More sophisticated discounting techniques need to be used to provide a net present value of the carbon offset during a whole rotation period, allowing for comparison with other offset projects (Moura-Costa 1995). Note, however, that the first contract between ICSB and NEP did not fully compensate ICSB for the incremental costs of RIL, especially with respect to foregone timber revenue. Hence, the commercial price of RIL as an offset is still evolving.
Figure 1. Carbon offset by the use of reduced impact logging (RIL) techniques
Cpf= carbon stored in primary forest;
Cf = carbon fixed at a certain point in time tf;
CL = total carbon fixed during the period from logging until tf, calculated by integration of the difference between the two curves.
Other Impacts
Apart from the direct benefits to the forest derived from RIL techniques, the project has had a broader impact. One of the most pleasing consequences of the project has been the development of a sense of pride among the logging crews and field staff. Furthermore, the training of a few logging crews has triggered a positive catalytic effect on the logging attitude around the region. Professional pride and competition has led to an unexpected improvement of the performance of other logging crews operating in the ICSB's concession. A simplified version of the RIL guidelines (adopting all the measures which do not require extra funding) was devised by the Regional Forest Manager of the Lahad Datu region (G. Mosigil, pers. comm.), and has been implemented throughout this part of the ICSB's concession.
Another positive effect of the project regards raising of awareness. The project has received substantial attention by local and international media (e.g., Bangkuai 1992, Parrish 1992, Anon. 1993, Soutif 1993, Wong 1993, Miller 1994, Vatikiotis 1994, Spaeth 1995), bringing to the attention of the general public that there are hopeful ways to conduct logging. For the specialised public, the project has served as a demonstration area of better logging practices, and attracted hundreds of local and international visitors since its inception. It has also been the target of research projects conducted by foreign and local scientists, both by individual projects (e.g., Nussbaum 1995) or field courses (e.g., Putz 1993).
At the international level, the project attracted the attention of institutions such as CIFOR, ITTO (Arentz 1994), the IPCC (Brown et al. 1995), and the GEF, which became interested in applying the project as a basis for spreading the use of RIL techniques throughout the region. The project has also served as an example of a successful model of joint implementation for carbon sequestration among the American utilities (Kinsman and Trexler 1995), and was the focus of an international workshop on joint implementation organised by the Harvard International Development Institute (Panayotou et al. 1993).
CONCLUSIONS
The first phase of the ICSB-NEP RIL project has created a positive momentum in the direction of achieving more sustainable logging practices. It has demonstrated that technical impediments can be overcome without major difficulty. The contract has been renewed and a second phase will be initiated early 1996.
Perhaps one of the major contributions of the project was to demonstrate an alternative model for funding tropical forestry. Carbon offsets in tropical forests, by reduced carbon loss or by increased carbon sequestration, as in reduced impact logging programmes, are splendid examples of global sharing of the financial burden of conservation. From a climate-change perspective this is eminently reasonable because CO2 and other heat-trapping gases circulate globally. From a political perspective carbon offset programmes should be acceptable in the tropics and elsewhere because they provide a mechanism for motivating the wealthy countries of the world to pay for a benefit of forest conservation that transcends national borders. It promotes the transfer of funds from industrialized countries to tropical countries as a commercial transaction, as opposed to charity (Marsh 1992, 1993).
It must be mentioned here that forestry will never be a primary solution for all the environmental problems related to the accumulation of greenhouse gases in the atmosphere. As pointed out by many countries during the Earth Summit, polluters must also attempt to reduce their emissions at source (Grubb et al. 1993). However, there are enough areas under forest cover or else deforested lands which, if properly managed, could sequester up to 15% of global emissions of carbon dioxide and other GHGs (Brown et al. 1995). Recognition of this potential may play a central role in improving the health and productivity of forests around the world.
ACKNOWLEDGEMENTS
I would like to thank R. Nussbaum and J. Tay for their comments and suggestions on this paper.
REFERENCES
Anon., 1993. A better way to cut trees. AsiaWeek, April 28 1993, pp. 49-50.
Arentz, F. 1994. Better logging is possible. Editorial for the ITTO Tropical Forest Update 4(3):1.
Bangkuai, J., 1992. Sabah's joint venture with Boston firm. Article about the RIL project in the New Strait Times, KL, 8 July 1992.
Blakeney, K.J. 1992. Environmentally friendly helicopter logging in Papua New Guinea. In: Proceedings of the International Symposium on Harvesting and Silviculture for Sustainable Forestry in the Tropics. Kuala Lumpur, October 1992.
Brown, S.; Cannell, M.; Heuveldop, J.; Kauppi, P.; Sathaye, J.; Singh, N.; Weyers, S.; Dixon, R.; Grainger, A.; Leemans, R.; Moura-Costa, P.H.; Nilsson, S.; Pinard, M,; Schopfhauser, W.; Sedjo, R. and Trexler, M., 1995. Chapter III.F. Establishment and management of forests for mitigation of greenhouse gas emissions. In: Working group II, Intergovernmental Panel on Climate Change, 1995 Assessment for the Framework Convention On Climate Change.
Cedergren, J., Falk, J., Garcia, A., Goh, F. and Hagner, M. 1994. Reducing impact without reducing yield. ITTO Tropical Forest Update 4(3):11-12.
de Graff, N.R. 1986. A silvicultural system for natural regeneration of tropical rain forest in Suriname. Agricultural University Wageningen, the Netherlands. 250 pp.
Dykstra, D.P. and R. Heinrich, FAO 1995. Code of forest practice harvesting. FAO Rome, 105 pp.
Grubb, M., Koch, M., Munson, A., Sullivan, F. and Thomson, K. 1993. The Earth Summit agreements. A guide and assessments. Earthscan Publications Ltd., London. 180 pp.
Froehlich, H.A., D.E. Aulerich, and R. Curtis. 1981. Designing skid trail systems to reduce soil impacts from tracked logging machines. Research Paper 36, Forest Research Lab, Corvallis, Oregon.
Hendrison, J. 1990. Damage-controlled logging in managed rain forest in Suriname. Agricultural University, Wageningen, The Netherlands.
Houghton, R. A., R. D. Boone, J. R. Fruci, J. E. Hobbie, J. M. Melillo, C. A. Palm, B. J. Peterson, G. R. Shaver, G. M. Woodwell, B. Moore, D. L. Skole, and N. Myers. 1987. The flux of carbon from terrestrial ecosystems to the atmosphere in 1980 due changes in land use: geographic distribution of the global flux. Tellus 39B: 122-139.
Jonkers, W.B.J. and H. Mattsson-Marn 1986. Logging damage in tropical high forest. FAO, Forestry Development Project, Kuching, Sarawak, Malaysia.
Jonkers, W.B.J. 1987. Vegetation structure, logging damage and silviculture in a tropical rain forest in Suriname. No 3 Ecology and management in rain forest in Suriname (Series). Wageningen, the Netherlands.
Kusoff, Kamaruzaman 1991. A survey of soil disturbance from tractor logging in a hill forest in Peninsular Malaysia. Malaysian Forestry and forest products research. Proceedings of the conference. Kuala Lumpur, pp. 16-21.
Kinsman, J.D. and Trexler, M.C. 1995. Into the wood. Electric Perspectives 20: 26-37.
Marsh, C., 1992. CO2 offsets as a potential funding source for improved tropical forest management. Tropical Forest Management Update 5 (2):15 and 7(2):2.
Marsh, C., 1993. Carbon dioxide offsets as potential funding for improved tropical forest management. Oryx 27(1):2-3.
Marsh, C. and Greer, A.G., 1992: Forest land-use in Sabah, Malaysia: an introduction to the Danum Valley. Philosophical Transactions of the Royal Society of London B, vol. 335, pp. 331-340.
Miller, M., 1994. Bringing back the rainforest. The San Diego Tribune, 2 and 9 February 1994. 10 pp.
Moura-Costa, P.H. 1995. Tropical forestry practices for carbon sequestration. In: Dipterocarp Forest Ecosystems: Ecology, Sustainable Management and Products, Schulte, A. (Ed). Indonesian-German Project, GTZ, Samarinda.
Nicholson, D.I. 1958. An analysis of logging damage in tropical rainforest, North Borneo. Malayan Forester 21:235-245.
Nussbaum 1995. The effects of selective logging of tropical rainforest on soil characteristics and growth of planted dipterocarp seedlings in Sabah, Malaysia. PhD thesis. Exeter University, UK.
Nussbaum, R., Anderson, J. and Spencer, T. 1995. In: Ecology, conservation and management of Southeast Asian rainforests. Primack, R.B. and Lovejoy, T. (Eds.). Yale University Press.
Palmer, J., and T.J. Synnott. 1992. Pages 337-373 in N.P. Sharma, editor. Managing the world's forests: looking for balance between conservation and development. Kendall/Hunt Publishing, Dubuque, Iowa.
Panayotou, T., Rosenfeld, A. and Kouju, L., 1994. To offset or not to offset: US power utility offsets CO2 emissions by financing reduced impact logging in Sabah. A case study for the 1994 Harvard International Institute of Development Asia Environmental Economics Policy Seminar, Bali 1994.
Parrish, M. 1992. L.A. firm helps utility with innovative plan. Los Angeles Times, 4 August 1992. 2 pp.
Pinard, M.A. 1995. Carbon retention by reduced impact logging. PhD thesis, University of Florida, Gainesville.
Pinard, M.A. and Putz, F.E. (in press). Retaining forest biomass by reducing logging damage. Biotropica.
Pinard, M.A., Putz, F.E., Tay, J. and Sullivan, T., 1994. Offsetting greenhouse gas emissions from New England through improved forest management in Malaysia. Journal of Forestry (submitted).
Putz, F.E. (Ed.), 1993. Proceedings of a course on research methods for silviculture and ecology, Forest Research Centre, Sandakan, June 1993. Forest Research Centre, Sabah, 250 pp.
Putz, F.E., and Pinard, M.A. 1993. Reducing the Impacts of Logging as a Carbon-Offset Method. Conservation Biology 7(4): 755-757.
Sabah Forest Department 1989. Forestry in Sabah. Sabah Forest Department, Sandakan, Sabah, Malaysia, 150 pp.
Soutif, M. 1993. Le réveil de la jungle. Geo (France) 177:8-21.
Spaeth, A. 1995. Angels in the rain forest. Time Magazine, April 3, 1995.
Vatikiotis, M. 1994. For profit's sake. Western utilities promote conservation in Sabah. Far Eastern Economic Review, April 14 1994, pp 68.
Wong, S.L. 1993. Green money, rewarding forestry. The Star (Kuala Lumpur), 6 and 13 October 1993. Environment Section.
by
Bryce J. Stokes, USDA Forest Service, Auburn, Alabama, USA
Niro Higuchi, Antonio C. Hummel, and Joberta V. De Freitas,
National Research Institute of Amazon (INPA), Manaus, Amazonia, Brazil
Jorge R. Malinovski, Federal University of Paraná, Curitiba, Paraná, Brazil
Abstract
Várzea forests are annually or periodically flooded. In the Brazilian Amazon region, these forests cover only about 5.5 million ha, but they provide a significant amount of wood for industry use. In the procurement system studied, a few wood buyers work through intermediaries to procure logs from thousands of small part-time producers. The payment system includes advances of food and supplies that are repaid with logs. Chainsaws are used to fell selected trees during the dry season (August to October). In the wet season (January to March), the logs are extracted by boats and moved to collection points. Large rafts or floats, are pulled to major cities from March to July.
Average tree diameter at stump height was 1.1 m, average height was 44.4 m, and average log volume was 11.1 m3. A felling crew cut and processed approximately 29 m3 per productive hour. Float sizes were 5,000 to 6,000 m3. The typical time to complete the procurement process was about 9 months. The average area of impact associated with tree felling was almost 0.09 ha. Corridor widths of 3 m were cleared during extraction. These impacts could be significant to regeneration, but more information is needed to fully assess the sustainability of this system.
INTRODUCTION
Brazil covers almost 65% of the Amazon River territory, 90% of which is still virgin forests. Dense tropical moist forest covers 360 million ha in the Brazilian Amazon. Improved silvicultural and utilization practices may be needed to satisfy increased demands, to improve economic development, and to ensure future access to forest products in the tropics. More technical information is required for sustainable forest management.
Proper forest operations, especially logging techniques, are needed to maintain and even improve site productivity. Hutchinson (1987) suggests that selectively logged broadleaved forests should have special protection and advance growth standing in residual and secondary forests. Improper harvesting practices can degrade forests so that future timber and non-timber values may be substantially reduced (Dykstra and Heinrich, 1992). Sustainable tropical forest management requires economically and environmentally acceptable harvesting practices.
There have been a few studies about the impacts of harvesting on the environment in terra-firme (non-flooded) forests (Lopes, 1987 and Veríssimo et al. 1992). A study by Oliveira (1992) evaluated only the impacts of traditional várzea (annually or periodically flooded) logging on forest succession. In the State of Amazonas, where almost all timber-based industries depended on várzea forests, no one has yet evaluated all harvesting operations impacts on the várzea ecosystem. More information is needed on the wood procurement system to assess its impacts on the entire Amazon basin.
BACKGROUND
The most important forest types in the Amazon River basin are dense tropical moist forests on terra-firme which cover an area of 330 million ha, and várzea forests which cover 5.5 million ha. Terra-firme forests have more above ground biomass and volume than várzea forests, which are characterized by a clean and open understorey. Typical commercial species of the várzeas forest are Virola surinamensis, Iryanthera surinamensis, Hura crepitans, Ceiba petandra, Carapa guianensis, Calophyllum braziliense, Copaifera sp., Naucleopsis caleneura, Pseudobambax munguba, Piranhea trifoliata, Hevea braziliensis, and Manilkara amazonica.
In 1989, the total Amazonian region produced 47.5 million m3 (IBGE, 1992) in roundwood. Hardwood veneer-plywood combination production is the most important forest industry activity in the Amazonas State. Hummel (1993) reported 116 timber-based industries in the Amazonas State (106 sawmills and 10 veneer-plywood combination industries). These two types of industries consumed equal quantities of raw materials. Of the 1992 total roundwood production, 90% came from várzea forests and 10% from terra-firme ones.
Little information is known about the economics or the environmental impacts associated with the current wood procurement system. Technical and operational aspects that are not very well quantified include productivity, costs, recovery, utilization, hazards to workers, ecosystems, waterways, possible depletion of the resource, destruction of regeneration in opening access corridors, and tree felling.
OBJECTIVES
This study assessed the traditional wood procurement system of the Amazon várzea forest to determine environmental impacts to forest sites and waterways and evaluated production, economics, and utilization standards. The approach was twofold: first to conduct a detailed case study of a conventional, representative wood procurement organization to learn about wood buying, harvesting, and transportation methods; and second, to assess stand structure changes, utilization, merchantable wood recovery, log specifications, damage to regeneration and remaining trees, and site disturbance during the harvesting.
METHODS
Information on locating, purchasing, and inventorying stumpage and on marketing forest products came through interviews with one of the roundwood buyers of the Amazonas State. Supplementing this information was an in-depth review of the number, type, and description of the producers.
Field observations took place at a typical várzea harvesting site on the Purus River, near the Lábrea municipality in the Amazonas State (Figure 1). The two observation periods took place during the felling (October 5 to 27, 1992) and extraction (March 19 to 31, 1993) phases. Forest inventory of the entire harvest area had already produced estimates of volume, species, and quality, as well as measurements of productivity and associated parameters for merchantable trees. The inventory method was to randomly distribute 20 sample plots (20 by 150 m each) within the study area and to evaluate the floristic composition and the structure of the natural forest by measuring all trees with dbh greater than 20 cm. The inventory crew collected botanical materials for scientific identification of the main tree species, and used the equation developed by Higuchi and Ramm (1985) to estimate the volume of standing trees.
During the first phase, records were kept of the time spent on cleaning around the selected tree, buttress cutting, cutting the main stem, pausing for maintenance, and moving from one tree to another. The gap area was measured and 18 trees were selected for measurements of total height, merchantable height, stump height, crown diameter, merchantable stem, diameter of the finest extremity, and diameter of the thickest extremity. Merchantable volume was calculated using Smalian's formula and compared to the buyer's volume table. Subplots were established within the site, in the access corridors, and under the logging slash to provide data on regeneration in the juvenile (5cm<dbh<20cm) and infantile (dbh<5 cm) stages.
The second phase monitored the extraction process. The sequences of this operation include: searching for logs felled during the dry season, log release from cut crown and stump, transportation of logs from the woods to a lake or to the river margin, and raft construction.
RESULTS
Wood Procurement System
The wood procurement system in várzea forests is a network of forest industries, buyers, and suppliers. Almost all forest industries depend on buyers and suppliers because few are inclined to do their own logging (Bruce 1989). Until 1983 the industries financed wood production, but today buyers are entirely responsible for the industry supply. There are no formal contracts or agreements among industries, buyers, and suppliers.
Almost 75% of the total roundwood consumed by timber industries of the Amazonas State is supplied by three buyers, one of whom was interviewed for this study. The buyer deals with his intermediaries or prepostos (for this season he used 18 in different areas) who in turn deal with ribeirinhos (wood cutters). The buyer advances money to the prepostos for boat and logging equipment maintenance, sustenance, and goods to exchange for roundwood produced by ribeirinhos. In 1993 the buyer expected to produce more than 150,000 m3 of raw material for veneer-plywood production.
Some prepostos act not only as intermediaries between the buyer and ribeirinhos, but also as loggers. In this case study one of the buyer's prepostos was studied. This preposto is both a logger and extractor. He prepares and tows logs to a lake or deeper river to bunch them in a raft, and he also supervises the raft transportation to Manaus. The preposto worked during the whole 1993 crop season on the buyer's property and used 33 ribeirinhos for logging and extracting. He had a small motorboat (33 HP) and a crew of four. He has worked for the buyer since 1989.
The preposto advances food and other supplies (such as medicines, tobacco, sugar cane rum, and fuel) to ribeirinhos and is repaid with roundwood. Very seldom can ribeirinhos be released from debt, which usually carries over from one year to the next unless production is outstanding.
Ribeirinhos are the basis of the roundwood production chain. Most work for the prepostos only during the felling and extracting phases. For the remaining six months of the year they have to do something else to support themselves such as harvesting rubber and Brazil nuts. Without financial help the ribeirinhos do not work. Although the preposto contacts only the ribeirinho, the whole family (even children) is involved in the work because the operation requires that the ribeirinho's to be away from home for long periods.
In this case study the ribeirinhos worked in the same area where the preposto did during the whole season. They cut selected trees, and they helped float logs out to the lake or deeper river and bunch them together in a raft. To identify the production of individual ribeirinhos, they print their initials on the logs. For the 1993 season the ribeirinhos contributed almost half of the preposto's production. Productivity of ribeirinhos ranged from 3 to 130 trees.
Forestry Inventory
The mean standing volume of trees greater 20 cm averaged 140 m3/ha (based on 20 observations) for all tree species, but only 43 m³/ha for potential commercial species. Three species - Sumaúa (Ceiba petandra Gaertn _ Bombacaceae), Muiratinga (Naucleopsis caloneura Ducke _ Moraceae) and Copaíba (Copaifera sp. _ Caesalpiniaceae) - contributed 55% of the total commercial volume. For the regeneration stage, defined as dbh less than 20 cm, the rarity of these three species suggests that they may be absent in the next rotation, and that production may not be completely sustainable for all species.
Felled Tree Volumes
The buyer's table estimates of merchantable volume averaged 74% of Smalian's volume, meaning that the ribeirinhos always got 26% less volume than what was calculated using Smalian's.
Harvesting
Three species contributed almost 100% of the 1993 harvest: Sumaúma, 46%, Muiratinga, 33%, and Copaíba, 20%. The felling crew used the following criteria in selecting individual trees for harvesting:
_ The flood mark on the selected tree was at least 3 m high to ensure sufficient water for the tree to float in the next wet season.
_ The estimated diameter of the selected tree was at least 50 cm.
_ The selected tree had good stem quality and a cylindrical and straight stem without apparent defects (decay or hollow).
_ The access-trail construction from the felled tree to the lake or river margin would be minimal.
The felling crew consisted of one chainsaw operator (chief) and two helpers to handle tasks such as cleaning around the tree to be felled and moving supplies. Aside from waiting 20 to 30 days before cutting crowns from felled trees, the crew took few safety precautions. Nobody had ever used safety equipment or been trained in safe felling practices. The crew could fell up to 21 marked trees per day with each tree yielding one log. The stump height (Figure 2) was always above 1.5 m, but could be higher depending on the size of buttresses. Directional felling was not always practical because of hidden climbers in the crown. Felled trees averaged 44.4 m high and 1.13 m dbh.
The average time to cut down one tree was 21.6 minutes. Effective activities (movement from tree to tree and felling) accounted for 58% (or 12.6 minutes) of the tree felling operation. The crew spent the remaining time _ 9.0 minutes or 42% _ in general activities (casual, personal, and waiting time). Productivity was 29 m3/hour of effective work.
The impacts caused by this operation covered a large area, but the effects on natural regeneration were not assessed. The gap size, estimated by crown projection to the ground, averaged 845 m2 _ twice as large as the average on the terra-firme forest (Vieira and Higuchi, 1991).
Extraction
During the wet season a team of six workers in a small canoe located logs that were felled during the dry season. Their first step was to release the log from the cut crown and stump with hand-saws and machetes. Most of the time, floatation is hindered by logs getting entangled with branches and climbers or attached in the bottom. These situations sometimes required diving or towing to release the log. After the release, one or more of the team was designated to move the log with the help of a cambito (wooden stick used to paddle the floating log) to the main trail where the motorboat was anchored, as illustrated in Figures 3-5. This was the most time-consuming part of the extraction operation.
In this study the motorboat pulled a raft with 18 pairs of logs averaging 15 m in length and connected by steel cable (Figure 6). The result was a raft of 200 m or more that the motorboat towed through the trail to a holding area in a lake or river deep enough to accommodate larger boats and rafts of 1,000 logs (usually 5,000 m3). The three common types of rafts (Figure 7) are:
_ Espinha de Peixe (fish-bone) is adequate for short distances and calm rivers. A steel cable connects logs in the center of the raft. This raft uses less steel cable and is therefore the cheapest, but it is not maneuverable enough for long distances or safe enough for full-flowing rivers.
_ Pente (comb-shaped raft) is recommended for long distances on full-flowing rivers. This raft is symmetric and can be towed from any side, especially valuable in tidal areas. This raft is the most expensive.
_ Cabeça (cone-shaped raft) is recommended for long distances on calm rivers.
The buyer usually takes responsibility for the towing operation. In this study the preposto transported the raft from Lábrea to Manaus, using his motorboat and the buyer's 340-HP towboat on the Purus River and adding a towboat when they reached the Solimões-Amazonas River. In the larger river, the towboat pulled additional rafts of many different sizes and types.
The raft took 20 days to reach Manaus, traveling day and night. This non-stop trip was a success because they did not have any accidents, thefts, or lost logs.
DISCUSSION AND CONCLUSIONS
The procurement system in the várzea forests of Brazil is pyramidal: a few buyers work through intermediaries or agents to buy wood from thousands of individual workers. A barter system predominates, in which buyers advance food and supplies to workers and are repaid through the felling and preparing of trees for extraction in the dry season. The following wet season, workers float the trees to a staging area and form them into rafts. The buyers transport the rafts to processing facilities at major cities.
Field visits were made and observations conducted on the felling, extracting, and rafting functions. Tree diameter at stump height averaged 1.1 m, height averaged 44.4 m, and log volume averaged 11.1 m3. A felling crew cut and processed approximately 29 m3 per productive hour. Tree felling time averaged approximately 22 minutes. Float sizes were 5,000 to 6,000 m3. In the study, raft transportation to Manaus took 20 days traveling 24 hours per day. The typical time to complete the entire procurement process was about nine months.
The intermediary or preposto in this study used 33 workers or ribeirinhos. Only three could not pay their debt. The buyer's volume table usually underestimated the calculated volume by 26% when using Smalian's formula. The difference may be an error in applying the formula or an underestimation on the part of the buyer's table. Additional research is needed to determine the source of the difference.
The average area of impact associated with tree felling was almost 0.09 ha. Corridor widths of 3 m were cleared during extraction. These impacts could be significant to regeneration; more information is needed to assess the sustainability of this system.
Forest management in várzea needs to be improved for stand succession, replacement, or sustainability. Although saplings and new growth responded positively to the gaps left by logging, there is little evidence that the most valuable commercial species will regenerate. Additional research is needed to recommend harvesting and regeneration improvements that will guarantee sustainability in producing raw material for timber-based industries in the Amazon region.
REFERENCES
Bruce, R.W. 1989. Log supply in Amazonas State, Brazil: Availability and constraints to utilization. Relatorio Tecnico para a ITTO (International Tropical Timber Organization). Fotocopiado.
Dykstra and Heinrich. 1992. Sustaining tropical forests through environmentally sound harvesting practices. Unasylva 43(169):9-15.
Higuchi, N. and C.W. Ramm. 1985. Developing Bole Wood Volume Equations for a Group of Tree Species of Central Amazonia (Brazil). Commonwealth Forestry Review, 64(1):33-41.
Hummel, A.C. 1993. Situação da Indústria Madeireira no Estado do Amazonas (1992). Technical Report of SEBRAE/IMA. Photocopied.
Hutchinson, I. 1987. Improvement Thinning in Natural Tropical Forests:Aspects and Institutionalization. In: Natural Management of Tropical Moist Forests, edited by F. Mergen and J.R. Vincent. Yale University.
IBGE (Fundação Instituto Brazileiro de Geografia e Estatística). 1992.Anuário Estatístico, Capítulo 44: Extração Vegetal e Silvicultura.
Lopes, C.A.C. 1987. Situação Atual do Programa de Pesquisa da SUDAM em Curuá-Una. In: Proceedings of "Encontro sobre Silvicultura e Manejo Florestal na Amazónia." pp.33-46.
Oliveira, M.V.N. d'. 1992. Exploração de Madeira em Várzea pelo Método Tradicional no Paraná Abufari no Médio rio Purus. EMBRAPA Boletim de Pesquisa, 7, 15p.
Veríssimo, A., P. Barreto, M. Mattos, R. Tarifa and C. Uhl. 1992. Logging Impacts and Prospects for Sustainable Forest Management in an Old Amazonian Frontier: the case of Parajominas. Forest Ecology and Management, 55:169-199.
Vieira, G. and N. Higuchi. 1990. Efeito do Tamanho de Clareira na Regeneração Natural em Floresta Mecanicamente Explorada na Amazônia Brasileira. In: Proceedings of the 6th Brazilian Forest Congress. pp. 666-672.
Figure 1. Map of the study area
Figure 2. Felling operation of copaíba (copaíba-particular species of tree
Figure 3. Preposto's 33 HP motorboat
Figure 4. Logs being removed from the flooded woods to the river margin
Figure 5. The entire transportation process from the woods to a lake (cambito _ a stick used to row a log from the stump to the principal route; madeira estocada _ stocked wood; lago _ lake).
Figure 6. Example of a"fish-bone" raft
Figure 7. Types of rafts: A. fish-bone; B. parallel strips or comb-shaped; C. head or cone-shaped.