reviews - University of Oxford

Transcription

International Forestry Review 2(1), 2000
INTERNATIONAL FORESTRY REVIEW (incorporating the Commonwealth Forestry Review)
The journal of the Commonwealth Forestry Association, c/o Oxford Forestry Institute, South Parks Road,
tel +44 (0) 1865 271037, fax +44 (0) 1865 275074, email [email protected]
Oxfrod OX1 3RB, U.K.
ADVICE TO CONTRIBUTORS
GENERAL
Manuscripts offered for publication should normally be
submitted in hard copy only, in the first instance, three copies
being sent to the Editor, International Forestry Review, c10
Oxford Forestry Institute, South Parks Road, Oxford OX1
3RB. Authors are advised to keep a copy of their manuscript
for reference, manuscripts will not be returned. The text
excluding tables, references and appendices should not
exceed 4000 words. Contributions must be original and not
have been submitted for publication elsewhere. The layout
of the text and style of table and figure legends and references
should conform to that of the Review (see Typography
below). The typescript should have a left hand margin of at
least 3 cm.
SUMMARY
A summary of not more than 150 words must be supplied,
together with 5 proposed keywords.
PREPARATION FOR PUBLICATION
Contributions will be referred to at least one expert referee.
Authors will be consulted if the paper is considered suitable
but alterations are thought desirable. After alterations have
been agreed and incorporated, the paper will be considered
final.
At this stage the paper should be submitted in electronic
form together with one hard copy. Discs should be written
in Wordperfect 6.1 1 8 or Word (6 or 97). Computer-drawn
graphics should be prepared using EXCEL, with separate
legends. If line drawings are supplied these should be in
black ink on white paper. Authors are asked to ensure that
symbols on figures are of sufficient size to be readable if
reduced for purposes of printing.
Photographs must be of a high quality, in black and white,
unless reproduction in colour has been agreed with the
Editor.
Page proofs will sent to authors for checking before
publication. These proofs are for the purpose of correcting
printing errors; only exceptionally and at the Editor's discretion
will authors be allowed to change their text at this stage.
New paragraphs should be indicated by a line space,
indents should not be used. Left justification should be
adopted throughout. Hyphenation should not be imposed on
the text.
Figures and tables should be supplied on separate sheets.
The use of graphics programmes and table editors should be
avoided. Captions should also be listed on a separate page
be written in italics with the words 'FIGURE'and 'TABLE'in
small capitals.
Authors are particularly asked to adopt the following
conventions for REFERENCES.
For text citations, papers should be referred to as (Smith
1998) and papers by the same author in the same year should
be distinguished by lettering in sequence (1998a, 1998b,
etc.) Where papers are by not more than three authors all
names should be cited, otherwise the first name followed by
'et al.'
At the end of the paper, the list of REFERENCES must
be arranged in alphabeticalordering without serial numbering.
The following standard forms of citation must be used:
author's name in small capitals after initial letter, all authors'
initials to follow surname, journal and book titles in italics,
journal volume number in bold. Second and subsequent lines
should not be indented.
Examples:
Journal paper: BRAITHWAITE,
L.W. 1986 Forests and fauna.
Forestry and Timber 22: 23-26.
Paper or chapter in proceedings: BOLAND,
D.J. 1986
Selection of species and provenances for tree introduction.
In: TURNBULL,
J.W. (ed.) Multipurpose Australian trees and
shrubs: lesser known species for fuelwood and agroforestry.
ACIAR Monograph No 1. 316 pp.
Book: PHILIP,MS. 1994 Measuring trees and forests. 2nd
edition, CAB International, Wallingford, England. 310 pp.
UNITS OF MEASUREMENT
Metric units are preferred and S.I. units should be used
wherever appropriate.
SCIENTIFIC NAMES
The complete scientific name (genus, species, authority and,
where appropriate, cultivar) should be cited at the first occasion
of mention of an organism and written in italics (authority in
normal type). If vernacular names are used, they must be
accompanied by the correct scientific name at fmt use.
TYPOGRAPHY
OFFPRINTS
Authors are asked to follow the conventions adopted in the
Review.
The hierarchy of headings is: CAPITALS, bold lower
ease, italics Idwer case.
10 copies of offprints together with a copy of the Review
will be provided free of charge to the senior author.
Additional prints will be supplied at cost.
SPECIAL ISSUE: REDUCED IMPACT LOGGING
Contents
Editor's introduction
REVIEWS
PAPERS
R. JEFFERY
and N. SUNDAR,
N. (eds.)
A new moral economy for India's forests?
Discourses of community and participation
J.E.M. ARNOLD
Reduced impact logging in the tropics:
objectives, principles and impact of research
P. SIST
Logging in South Cameroon: current methods
and opportunities for improvement
W.B.J. JONKERS and G.J.M. VAN LEERSUM
RIL for real: introducing reduced impact
logging into a commercial forestry operation
in Guyana
S. ARMSTRONG
Testing the applicability of reduced impact
logging in greenheart forest in Guyana
P. VAN DER HOUT
Lessons learned from the implementation of
reduced impact logging in hilly terrain in Sabah,
Malaysia
M.A. PINARD, F.E. PUTZ and J.TAY
Reduced impact logging as part of the
domestication of neotropical rainforest
N.R. DE GRAAF
Benefits, bottlenecks and uncertainties in
the implementation of reduced impact
logging techniques
D.S. HAMMOND, P. VAN DER HOUT,
D.S. CASSELLS, R.J. ZAGT, J. EVANS
and G. MARSHALL
J-P. LEONARD
Contribution h la typologie des principaux
systbmes forestiers. Essai de classification
physionomique des for&tsh partir des facteurs
sociaux generateurs. [Contribution to a typology
of principal forest systems. An attempt at a
classification based on social determinants]
J. GUILLARD
A.L. MITCHELL
and S. HOUSE
David Douglas - explorer and botanist
J. R. ALDHOUS
A. MILES
Silva: the tree in Britain
A.J. GRAYSON
M.F. NEWMAN,
P. F. BURGESS
and
T.C. WHITMORE
Malesian Dipterocarps. Foresters' CD-ROM
manual
S. HARRIS
THEROYAL
HORTICULTURAL
SOCIETY
Plantfinder (CD-ROM)
P.G.ADLARD
L. NSHUBEMUKI,
H.J.M. MWANSOKO
and
A.G. MUGASHA.
Istilahi za Elimumisitu Kiingereza-Kiswahili
[Forestry Terminology English - Kiswahili]
P.J.WOOD
COMMENT
Comment on paper by R.Tipper and
B. de Jong 'Quantification and regulation
of carbon offsets'
I. G. ENTING
MISCELLANEA
Deliberate introduction of species:
research needs
Carbon accounting
Reply to Enting
R.TIPPER and B. DE JONG
The World Forestry Center
The CFA's goal -
70 promote the well-being of the urorld's forests and those who depend upon them
i
ii
FORTHCOMING INTERNATIONAL
EVENTS
64
New corporate member
Obituaries
L. Roche
ASSOCIATION AFFAIRS
J.V. Thirgood
Chairman honoured
66
Farewell lunch to the Duke of Buccleuch
66
Formation of UK Branch
66
TRANSLATIONS OF SUMMARIES
French
Spanish
The International Forestry Review is a peer-reviewed journal.
international Forestry Review 2 ( l ) ,2000
1
EDITOR'S INTRODUCTION
This special issue of the Review contains selected papers
originally presented at an international workshop on reduced
impact logging (RIL) hosted by the Iwokrama Centre,
Guyana in April 1999.'
The Iwokrama International Centre for Rain Forest
Conservation and Development originated from the 1989
Commonwealth Heads of Government meeting in Kuala
Lumpur when the government of Guyana offered to make
some one million acres of rain forest available to the
international community for 'a project for developing and
demonstrating methods of sustainable management of
tropical rain forests and of conserving biological diversity'
(the words of the then President, President Hoyte quoted in
Kerr '). The Centre was created with the support of the
Commonwealth Secretariat, the Global Environment
Facility, the United Nations Development Programme,
Canada's International Development Research Centre and
other donors. Half the reserve of 360,000 ha reserve, which
lies 150 miles south of Georgetown, is retained as a
wilderness reserve to provide a reference standard on
ecological processes. The other half is managed for
commercial production, experimental research and nonwood producing activities such as eco-tourism
The Centre is constituted as an autonomous international
research and development centre governed by an
international Board of Trustees. A key objective of the
Centre's 10 year business plan is for the Centre to become
financially self-sufficient through a set of core activities
based largely on the endowment of the Iwokrama Forest.
The Centre became effectively operational in July 1998 with
the initiation of a Sustainable Human Development Project
funded by the U.K. Department for International Development. It now enjoys support from some 28 countries or
international agencies contributing cash or kind to its
operations. The Centre has grown rapidly with total staffing
now exceeding 70, including 15 forest scientists and resource
management professionals.
As part of its sustainable management programme, the
Centre hosted the RIL workshop in partnership with the
TROPENBOS Foundation and the Guyana Forestry
Commission. The purpose of the meeting was to share
experiences among rain forest management researchers and
practitioners with the aim of distilling knowledge of best
practices in logging. We are glad to collaborate with the
Centre's Director General, Mr David Cassells, in giving
wider access to material prepared for this important
workshop.
Iwokrama, an environmental agenda for the world. Commonwealth Currenrs l999 Issue 2: 4-6.
* Kerr. B. 1993 Iwokrama: the Commonwealth Rain Forest
Programme in Iwokrama. Commonw. FOKRev. 72(4):303-309.
l
International Forestty Review 2 ( I ) , 2000
3
PAPERS
Reduced-impact logging in the tropics : objectives,
principles and impacts
CIRAD-for& rad-For&, BP 5035, 34032 Montpelliec France
[email protected]
SUMMARY
The objectives and principles of the RIL techniques in the tropics are described and a review presented of the most important results of
research carried out on the impact of logging on the remaining stand and forest dynamics. The main objective of RIL techniques is to reduce
substantially disturbances to soil and residual vegetation in comparison with conventional logging. RIL is mainly based on close planning
and control of all harvesting operations. The amount of logging damage not only depends on the techniques used but also on logging
intensity. In Africa, logging damage is usually slight because harvesting intensity rarely exceeds 2 trees per ha. The extreme picture is found
in South East Asia where logging intensity averages 9 trees per ha; this involves damage which can affect half of the forest stand. With
such logging intensities, RIL techniques are not efficient in reducing damage significantly.
Keywords: logging damage, logging intensity, reduced impact logging.
INTRODUCTION
Since the late 1950s, due to increased use of heavy
machinery for timber extraction, the impact of logging on
tropical forests has attracted the attention of silviculturists
and forest managers. In addition, a growing awareness of
the need to protect forest ecosystems' functions and to
maintain biological diversity in production forest raises
the question whether timber harvesting can be compatible
with other forest production. Efforts towards sustainable
forest management, such as the ITTO year 2000 objective,
which aims to bring the forest estate under sustainable forest
management, have promoted the implementation of
Reduced Impact Logging techniques (RIL), also called LIL
(Low Impact Logging) or LIH (Low Impact Harvesting)
techniques. RIL's main objective is to reduce soil
disturbance, impacts on wildlife, and damage to residual
trees. RIL has been recently implemented and tested in
various tropical regions, particularly in South East Asia
and Latin America (Sabah - Pinard and Putz 1996,
East Kalimantan - Bertault and Sist 1995, 1997, Sist et al.
1998, South America - Hendrison 1990, Uhl and Veira 1989,
Johns et al. 1996, Bird 1998). In this context of increased
effort to achieve sustainable forest management, codes of
practice and RIL guidelines have been produced by forestry
research organisations such as FAO, CIFOR and CiradFor& (Dykstra and Heinrich 1996, Sist et al. 1998, Durrieu
de Madron et al. 1998) as well as national forestry
departments (e.g. Vanuatu Forestry Department, Sabah
Forestry Department).
This paper aims to present the main objectives and
principles of RIL techniques. Research on the impact of
logging on forest ecosystems is essential for the development
of silvicultural techniques, including logging, that are
compatible with sustainable forest management. Accordingly, the second part of the paper presents a synthesis of the
most important results of research carried out on three
continents (Africa, South America, Asia) into the impact of
logging on the remaining stand and forest dynamics.
REDUCED IMPACT LOGGING: OBJECTIVES AND
PRINCIPLES
Objectives
As set out by Pinard et al. 1995, the main objective of RIL
techniques is to reduce disturbances to soil and residual
vegetation by at least 50 % in comparison with conventional
logging. It is also expected that limiting the impact of logging
will result in protection and maintenance of the long-term
integrity and value of the forest resources and environmental
services they provide (non-wood forest products, wildlife,
endangered andlor rare species, watershed, soil protection
against erosion, etc.). RIL tech-niques must also be
economically attractive to concession-aires and, .desirably,
direct costs of logging operations should be significantly
reduced compared with conventional techniques.
4
P. Sist
Principles
tactical. Harvesting plans are medium-term plans which are
generally set up for about 5 years. All the information
included in the strategic plan (Figure 1) must be described
in a written document and depicted on 1 :25,000 scale maps.
One important task of the strategic plan is to define the
type of extraction, which is primarily topography dependent.
In areas where slopes are mostly less than 30%, ground
skidding is permissible. On slopes ranging between 30% and
70%, ground skidding should not be permitted because of
the extensive damage to both soil and vegetation which is
likely to result. In this range of slopes, skyline yarding
systems are an appropriate extraction system. Areas with
slopes over 70 % must be excluded from cutting and should
Forest management plans and planning of harvesting
operations
Success in reducing logging damage cannot be achieved
without close planning and control of harvesting operations.
Harvesting plans are generally included in a broader longterm (20 years) land-use forest management plan which is
designed to ensure sustainable utilisation of the forest. This
planning process takes into account the ecological, environmental and socio-economic features of the concession
(Figure 1). Harvesting plans are of two types: strategic and
FOREST MANAGEMENT PLAN
Long term plan ( 5 20 years)
Management for a long term sustainable utilisation of forest resource
Land use features for the identification of protection and production areas
Assessment of the environmental and socio-economic components of the concession
Broad scale planning map 1:50,000
STRATEGIC PLAN
Medium-term plan (5 years)
lI
Delimiting and mapping of the annual coupes, the protected and
production areas
Vegetation type classification and topography
Estimated volume production in each coupe
Harvesting systems and equipment
Training sessions schedule
Main road design
Maps 1 :25,000
I
..........................
...................
Tactical plan Year 2
.
..................
I
RIL PLANNING AND IMPLEMENTATION
Planning of logging operations at the annual coupe scale
Pre-harvesting forest inventory (100% timber inventory)
Planning of felling
Planning of secondary roads, landings and skidding trails
Supervision of logging operations
Planning of post-logging operations
Detailed tactical logging map 1:2,000 (see Figure 2)
FIGURE1 Forest management and harvesting plans
Source: Sist et al. (1998)
...................
Tactical plan Year n
.
..................
RIL: research objectives and impact
be identified as protection forest. The same holds true for
riparian zones and areas of unique forest habitat.
Tactical plans provide technical procedures and planning
detail for the harvesting operations to be carried out in the
annual coupe. A tactical plan generally covers three
successive phases, namely the pre-harvesting phase, the
logging operation phase and the post-logging phase.
Pre-harvesting operations
Pre-harvesting activities aim to collect all essential
biophysical data in order to plan the logging operations for
the annual coupe. The work consists mainly of forest
inventory and topography assessment. Pre-harvesting
operations lead to the construction of a tactical logging map
and a tactical logging plan document.
Growing stock survey
For RIL implementation, the minimum requirement in stock
survey is to record at least all the harvestable timber trees in
the annual coupe. For each tree to be harvested, it is
recommended that the following data be recorded:
tree number and local or commercial/trade name,
position in the cutting block on a map at a 112000 scale,
estimated diameter class above buttresses.
It is also recommended that records include the potential
crop trees (or PCT) which are commercial timber species
with dbh (diameter at breath height) below the harvestable
limit at the time of logging but likely to comprise the next
harvest. This will allow more accurate determination of
felling direction in order to protect as many as possible PCT
during logging. Protected tree species and important wildlife
resource trees must be also included in the inventory.
Climber cutting
Climbers can compromise seriously both feller safety and
directional felling. Therefore, all climbers more than 2 cm
in diameter that are attached to the canopy of harvestable
stems should be cut at least one year before logging. Ficus
spp. are excluded from this practice, given their value to
wildlife. Climber cutting should normally be done in conjunction with the inventory operations.
Topographic survev
Because road and skidding trail networks are planned and
designed according to the topography of the terrain, it is
essential to have accurate topographic maps. These may be
produced using the many available methods and tools (e.g.
aerial photos, satellite images, radar images). However, if
the remote sensing images are not sufficient to produce maps
at a suitable scale for logging planning (1:2,000), topographic maps must be prepared through an intensive field
survey. This survey is particularly recommended on steep
terrain such as hill forests (Sist et al. 1998). In very flat
5
terrain, topographic assessment might not be necessary and
only swamp areas, rivers and streams should be recorded
during growing stock survey.
Protected areas
All areas appearing to merit protection are noted during
growing stock survey andlor topographic survey, and
subsequently delineated on the tactical map. In these
protected areas, logging operations are not allowed.
Protected areas can be generally defined as follows:
Unworkable areas: areas that are too steep, rocky, and1
or have very little commercial timber,
Sacred areas: areas that have cultural or religious value
for the local residents. Sacred areas must be defined in
consultation with local populations and clearly marked
as protected areas on the logging plan maps,
Conservation areas: areas that preserve unique andlor
fragile habitats, and areas of high biodiversity, from any
human disturbance including hunting. These areas must
be representative of the different ecosystems occurring
in the concession and can only be defined through an
assessment of the biophysical features of the area which
should be designed within the strategic plan,
Stream buffer zones: areas adjacent to streams where
logging activities are restricted. Streams are considered
to be watercourses that flow for at least two months in
most years. Stream buffer zones vary in width from 20200 m according to the size of watercourse. Stream
buffer zones must be recorded during topographic survey
and drawn on the tactical plan maps.
Planning roads. landings and skid din^ trails
The location of major roads is decided as part of the
strategic plan while secondary roads are addressed in the
tactical plan. Major road design may be modified and
improved however, following the pre-harvest topographic
assessment carried out during tactical planning. In RIL,
forests roads must be constructed according to
environmentally sound engineering practices in order to
minimise soil erosion and stream sedimentation (Dykstra
and Heinrich 1996).
The density of landings (temporary log storage areas)
and their area can be limited by planning roads and landings
prior to logging, drawing upon knowledge of the topography
and of the spatial distribution patterns of harvestable trees
assessed in the pre-harvest inventory. To minimise the
environmental impacts associated with landings, the
following rules should be followed:
develop landings adjacent to roadways,
restrict their size to 0.2 ha (approximately 30 X 60 m),
avoid placing landings in a r m excluded from harvesting,
locate landings on ridges to ensure uphill skidding,
outside areas excluded from harvesting and on areas easy
to drain.
6
P. Sist
Skidding trails are planned according to the tree position
and terrain topography recorded in the pre-harvesting forest
surveys. The following rules serve as the foundation of any
skidding trail design:
ground skidding is not acceptable on slopes greater than
30%, nor in protected areas or stream buffer zones,
stream crossings are not favoured. Where they cannot be
avoided, crossing points must be clearly shown on the
map and must be approved after a field check by the
Planning or Forest Inventory officer of the Forest
Authority,
the skidding trail network must be optimised according
to the position and density of trees to be felled in order
to minimise the length of trails in the forest.
Planning of felling
The major objective of directional felling is to position
the stem so as to facilitate extraction of logs. The second is
to avoid damage to PCT (Potential Crop Trees). Directional
felling can also be used to protect individual trees of
endangered or rare species which have been recognised and
recorded during the pre-harvesting forest inventory. The
main guidelines for planning directional felling are:
the tree must be felled either toward or away from
skidding trails or cable ways at an oblique angle of
approximately 30" to the skidding direction unless the
tree can be felled directly on to the skidding trail,
where possible, trees should be felled in the direction of
existing canopy gaps,
on steep slopes, trees must be felled uphill unless their
downhill lean is too great,
trees within or near a stream buffer zone must be felled
so that the crowns fall outside the buffer zone,
proper felling procedures must be applied to avoid
splitting of the tree during felling and to minimise wood
waste.
Field experience shows that the best operators to decide
the direction of felling are the fellers~themselves.The
opening of skidding trails prior to felling operations is
recommended as this will help the feller to check the best
feasible direction of felling of each tree.
Tactical maps and written plans
Pre-harvesting inventories and planning must lead to the
development of a tactical map which includes all the
information needed to achieve the logging operations as
defined in the plan. The tactical plan must also include a
written document giving all the technical details of each
activity to be carried out (pre-harvesting, harvesting and
post-harvesting activities). The tactical map must be at a
large scale (1 :2000 is suggested); it gathers the following
information:
topographic contour lines if necessary (5 m interval or
less),
l
position of each tree to be felled, marked with its
inventory number,
road network and landing locations,
skidding trail network,
protected areas.
Harvesting operations
Felling (team, maps and material)
Felling operations must be carried out by skilled personnel
equipped with appropriate safety gear and using properly
maintained equipment. The fellers must be familiar with
directional felling techniques (see Klasson and Cedergen
1996). It is therefore important that the logging company
provides training for fellers who have no experience in these
techniques.
Skiddine trails - marking and o ~ e n i n g
During log extraction the following practices and equipment
are recommended:
skidding trails must be opened according to the planned
network shown in the tactical logging map and in the
forest,
skidding trail width must not exceed 4 m,
skidder operators are not allowed to leave the marked
trails without permission,
skidding trails are not allowed to cross streams wider
than 5 m or gullies. Where this must be done, crossing
must be made at a site where there is a rock base or a firm
base, and the stream bed must be protected with logs or
a temporarily culvert constructed,
wheeled or crawler skidders are to be preferred to crawler
tractors which should be used only in road construction
and maintenance,
the skidder must have a powered winch with at least 30
m of wire rope as well as an arch or other support for
suspending the end of the load off the ground,
the blade of the skidder should not exceed 3 m and
blading should be avoided as much as possible in order
to minimise the impact on soil,
skidding of logs behind the tractor is not allowed on
slopes over 30%; on such slopes log must be hauled up
the hill using the winch,
blading and log extraction during rainy days must be
avoided in order to limit soil erosion.
Payment system
RIL techniques require the employment of skilled, welltrained and responsible staff as well as a modified piecework
remuneration system in which payment takes into account
the quality of the work. The remuneration system used in
conventional logging, which only takes into account the
commercial volume produced, is not effective in motivating
the application of RIL techniques. A compensation system
that rewards workers for good practices is needed.
7
RESEARCH ON THE IMPACT O F LOGGING ON THE
FOREST
Post-hawesting operations
Rehabilitation of skidding-trails
After log extraction, cross-drains will be constructed on
skidding trails in order to limit soil erosion. The following
procedures are recommended:
cross-drains should be built at an angle of 60" to 80' as
measured from the longitudinal axis of the skid-trail,
the frequency of cross-drains increases with slope
(Table 1 sets out the scheme developed for use in Sabah),
any temporarily stream crossing structure must be
removed from the skidding trail.
TABLE
1 Minimum cross-drainfrequencyforskidding trails
and roads
Slope (%) on skidding trails
Cross-drain spacing
< 10
10-20
20-30
No cross-drain
30 m
20 m
Slope (%) on roads
Cross-drain spacing
<5
5-15
15-20
No cross-drain
120 m
80 m
Source: RIL guidelines in Innoprise, Sabah (Pinard et al. 1995).
Road closure
Secondary or minor roads which are not to be used until the
next felling cycle must be closed. Decisions of road closures
are the responsibility of the officer in charge of logging and
must be made in consultation with local stakeholders. Road
closure includes the removal of log culverts and temporary
bridges as well the construction of cross-drains according to
the guidelines listed in Table 1.
t
operations
Other ~ o s harvesting
Other important post-harvesting operations are:
control access to the permanent forest estate. Solely
those individuals directly involved with forestry
operations should be permitted into the permanent forest
estate. A possible exception would be to allow local
stakeholders access, however they would need to register
their activities with the forest officer. Only ephemeral,
subsistence activities should be allowed,
proper maintenance of road surfaces, roadside ditches,
cross-drains and stream crossings. Secondary or minor
roads can be closed off if they will not be used again
until the next felling cycle,
cleaning landings and temporarily camps: bark, branchwood, etc to be burnt, rubbish, including oil or fuel,
drums, wire rope to be removed, cans and other metallic
waste to be buried.
Most of the studies related to the impact of logging on forest
stand provide important information on the quantity and the
type of damage caused by logging according to the
harvesting intensity and the extraction techniques used
(Nicholson 1958, 1979, Jonkers 1987. Hendrison 1990, Uhl
and Veira 1989, Cannon et al. 1994, Johns et al. 1996,
Pinard and Putz 1996, Bertault and Sist 1997, Sist et al.
1998). More recently, growth modelling has also been
developed to predict the yield of production forest according
to the main parameters such as the level of cut and amount
of logging damage (Vanclay 1994, Ong and Kleine 1995).
Beside studies clearly focused on best silvicultural practices,
numerous experiments have been carried out on the impact
of logging on biodiversity, soil erosion and fauna (Johns
1991, 1992, Malmer and Grip 1990, Gullison and Hardner
1993, Thiollay 1992, White 1994). These studies are
important in our understanding of the functioning of the
complex tropical forest ecosystem. However, it is beyond
the scope of this paper to review the whole range of such
studies; instead attention is concentrated on the contribution of research in the identification of best silvicultural
practices.
There are still very few studies on the costs of RIL
implementation though this point is essential to convince
concessionaires and decision makers to promote the
technique. Compared with conventional techniques, the main
costs of RIL undoubtedly arise in the planning stage.
However, in the Brazilian Amazon, Barreto et al. 1998
demonstrated that proper planning increased labour
productivity and reduced waste during logging operation
resulting in a net financial benefit of US$3.7 per m'.
Damage to the stand caused by logging activities
Amount of damage according to regions
Based on studies carried out on the three continents (Africa,
America, Asia) where tropical rain forests occur, Table 2
presents general patterns regarding the amount of damage
according to the level of cut. Logging intensity (i.e. the
number of stem harvested per ha varies according to regions
and continents. In Africa, it is very low (1-2 trees ha-' or 1520 m' ha-') while the highest, in South East Asia, indicates
a mean of 8 harvested trees ha-' representing a volume of
between 80 to 100 m3 ha-l (Table 2). South America shows
an intermediate position with an average of 5-6 harvested
trees ha-'. However, timber tree species are not uniformly
distributed in the forest and logging intensity can therefore
show very high variations within the same locality. There are
sites in Africa and South America where logging intensity is
comparable to that recorded in South East Asia. In Guyana,
for example, the main commercial species known as
greenheart (Chlorocardium rodiei) is often found in clumps
in primary forest. This can result in alogging intensity rising
to more than 15 trees ha-' (Zagt 1998, Van der Hout this
l e
P
RILne 8
RILn > 8
CNV
l
FIGURE
4 Mean percentages (bars + standard deviation) of
trees damaged (injured or killed, % of the original tree
population) by felling and skidding in RIL and CL
(conventional logging) with different felling intensity,
Kalimantan
.,:
Notes: RIL
RIL with a felling intensity c 8 trees ha-',
RIL ?,: RIL with a felling intensity r 8 trees ha-'. Hatched
bars: felling damage, empty bars: skidding damage, black
bars: total damage
Source: Sist et al. (1998)
The main objective of directional felling is to position
logs for easier extraction. The effect of vine cutting in
reducing the felling damage seems to be dependent on their
density prior to logging (Cedergen 1996). In Malaysia, in a
forest with high climber density (376 ha-', dbh > 2 cm),
Appanah and Putz (1984) observed that vine cutting prior to
logging reduced the number of trees pulled down during
felling by approximately one-half. In contrast in Sabah, in a
forest with a lower climber density (189 ha-'), Cedergen
(1996) demonstrated that vine cutting had no effect in
reducing the felling damage. Felling damage intensity mainly
depends on biophysical factors such as the height of the tree,
the size of the crown and the topography (Cedergen 1996).
Techniques capable of significantly reducing felling damage
to stands are not yet available in the tropics. The only
method of reducing such damage that is currently available
is to limit the level of cut.
Impact of logging damage on long term forest dynamics
Logging clearly has a long term impact on the dynamics of
the forest. Tree mortality rate in logged-over forests is
significantly higher than in primary forest (3-5 % per year
vs.1-2%) for 5 to 10 years after logging (Nicholson 1979,
Primack et al. 1985, Manocharan and Kochumen 1987).
This can be partly related to the mortality of trees damaged
by logging which show a much higher mortality rate than
undamaged ones. In East Kalimantan, two years after
logging, mean annual mortality rate of damaged trees was
5.5 % versus 1.1 % only for undamaged (Nguyen-ThC et al.
1998). However, canopy opening also creates favourable
light conditions for some species, including commercial
9
ones, the growth of which is stimulated at least during the
three years following logging (Primack et al. 1985, Jonkers
1987, Nguyen-ThC et al. 1998). This is the case in South
East Asia where dipterocarps show a positive growth
response to extra light (Nguyen-ThC et al. 1998). In the last
case, a high level of logging damage has a negative impact
on forest recovery and hence on volume increment. The high
mortality of highly damaged forest is not compensated by
the growth of the remaining trees and the recruitment of new
trees with dbh r 10 cm. This last result suggests that forests
harvested using RIL may recover faster and produce more
growth than those logged under conventional practice. The
length of the felling cycle may be significantly reduced in
consequence. Indeed, the long-term benefit of RIL may arise
mainly in the reduction of the felling cycle that is feasible.
CONCLUSIONS AND DISCUSSION
Reduced impact logging must not be regarded as the one
single activity able to ensure the achievement of sustainable
forest management. RIL is only a technical and silvicultural
procedure which must be integrated into a defined management plan for long-term use of the forest. Certification
introduces a comprehensive discipline aimed at achieving
sustainable management and the adoption of RIL is a prerequisite for certification (Udarbe et al. 1994). In the tropics,
more and more logging companies are now aware of the
value of certification if they want to increase their share of
the international wood market. This context is therefore very
favourable for the promotion of RIL. More studies on the
cost of RIL implementation compared with conventional
techniques should be undertaken since the common belief of
loggers is that RIL techniques are more expensive.
The impact of logging and the intensity of damage are
directly linked to logging intensity and harvesting
techniques. Because logging damage in West and Central
Africa is usually very low (affecting less than 15 % of the
original stand), the reduction of such damage by 50 % will
be certainly less noticeable than in South East Asia where
logging impact is very high. However, RIL is not only a
technique to reduce the damage, it is also a procedure to
optimise resource utilisation through forest inventory and
planning of harvesting. In the tropical forest of Central and
West Africa, this last aspect of RIL is undoubtedly the most
important one and may have much more significant impact
on silviculture than the reduction of damage only. In Africa,
successive fellings at a very short intervals are very common
and often associated with poor, or a complete lack of, forest
inventory. Such practices are not compatible with sustainable management and could be easily avoided if there was
an accurate inventory of the forest resources before
harvesting. Although the mean logging intensity in Africa
and South America is usually much lower than that recorded
in South East Asia, this intensity is not uniform throughout
the forest. These variations are likely to have important
consequences in terms of silviculture (see Van der Hout, this
volume). Our knowledge on how to define sound silviculture
10
P.Sist
adapted to local variations of logging intensity is still poor
and more research is required on this matter.
REFERENCES
APPANAH,
S. and PUTZ,F E . 1984 Climber abundance in virgin
dipterocarp forest and the effect of pre-felling climber cutting
on logging damage. Malay. For. 47 (4): 335-342.
BARRETO,
P,, AMARAL,
P., VIDAL,E. and UHL,C. 1998 Costs and
benefits of forest management for timber production in eastern
Amazonia. For. Ecol. Manage. 108: 9-26.
BERTAULT,
J-G. and SIST,P. l995 The effects of logging in natural
forests. Bois er forifs des tropiques 245: 5-20.
BERTAULT,
J-G. and SIST,P. 1997 An experimental comparison of
different harvesting intensities with reduced-impact and
conventional logging in East Kalimantan, Indonesia. For: Ecol.
Manage. 94: 209-21 8.
BIRD,N.M. 1998 Sustaining the yield. Improved timber harvesting
practices in Belize 1992-1998. Natural Resources Institute,
Chatharn, UK.
CANNON,
C.H., PEART,
D. R., LEIGHTON,
M. and KARTAWANATA,
K.
1994 The structure of lowland rainforest after selective logging
in West Kalimantan, Indonesia. For: Ecol. Manage. 67: 49-68.
CEDERGEN,
J. 1996 A silvicultural evaluation of stand
characteristics, pre-felling climber cutting and directional
felling in a primary dipterocarp forest in Sabah, Malaysia.
Doctoral thesis, Swedish University of Agricultural Sciences,
UmeB, 1996.
DURRIEU
de MADRON,
FORNI,E., MEKOK,
M. 1998 Les techniques
faible impact en for& dense humide camerounaise. Serie
FORAFRI, Document n o l7, 29 pp. Cirad-For&, Montpellier.
DYKSTRA,
D. and HEINRICH,
R. 1996 FAO model code of forest
harvesting practice. FAO, Rome, 85 pp.
GULLISON,
R.E. and HARDNER,
J.J. 1993 The effects of road design
and harvest intensity on forest damage caused by selective
logging: empirical results and simulation model from the
bosque chimanes, Bolivia. For. Ecol. Manage. 59: 1-14.
J. l990 Damage-controlled logging in managed
HENDRISON,
tropical rain forests in Suriname. Series on the ecology and
management of tropical rain forests in Suriname. Wageningen
Agricultural University, Wageningen, Netherlands, 204 pp.
JOHNS,
A.D. 1991 Responses of Amazonian rain forest birds to
habitat modification. J. Trop. Ecol. 7: 417-437
JOHNS,A.D. 1992 Vertebrate responses to selective logging:
implications for the design of logging systems. Phil. Trans.
Roy. Soc. London ( B ) 335: 437-442.
JOHNS,J.F., BARRETO,
P. and UHL,C. 1996 Logging damage
during planned and unplanned logging operations in the eastern
Amazon. For. Ecol. Manage. 89: 59-77.
JONKERS,
W.B.J. 1987 Vegetation structure, logging damage and
silviculture in a tropical rain forest in Suriname. Ph.D.thesis,
Wageningen Agricultural University, The Netherlands.
&ASSON,B. and CEDERGEN,
J. 1996 Felling the right way. Some
hints on the art and science of directional felling. I7TO Trop.
For Update 6 (3): 5-7.
MANOCHARAN,
N. and KOCHUMMEN,
K.M. 1987 Recruitment, growth
and mortality of tree species in a lowland dipterocarp forest in
Peninsular Malaysia. J. Trop. Ecol. 3: 315-330.
NICHOLSON
D.I. 1958 An analysis of logging damage in tropical
rain forests, North Borneo. Malay. For. 21 (4): 235-245.
N 1 c ~ o ~ D.I.
s o ~1979 The effects of logging and treatment on the
mixed dipterocarp forests of Southeast Asia. FAO, Rome.
Report FO: MISC/79/8, 65 pp.
NGUYEN-THE,
N., FAVRICHON,
V., SIST,P., HOUDE,L., BERTAULT,
JG. and FAUVET,
N. 1998. Growth and mortality patterns before
J-G. and KADIR,K. (eds.)
and after logging. In: BERTAULT,
Silvicultural research in a lowland mixed dipterocarpforest of
East Kalimantan, the contribution of STREKproject, CIRADFor& Publication pp. 181-216.
ONC,R. and KLEINE,
M. 1995 DIPSIM. A dipterocarp forest
growth simulation model for Sabah. Forest Research Centre
Research paper No. 2, Forest Department, Sabah, Malaysia, 94
PP.
PINARD,
M.A., PUTZ,F.E.,TAY,J. and SULLIVAN,
T.E. 1995 Creating
timber harvesting guidelines for a reduced-impact logging
project in Malaysia. J. For. 93 (10): 41-45.
PINARD,
M.A. and PUTZ,F.E. 1996 Retaining forest biomass by
reducing logging damage. Biotropica 28 (3): 278-295.
PRIMACK,
R.B., ASHTON,
P.S., CHAI,
P. and LEE,H.S. 1985 Growth
rates and population structure of Moraceae trees in Sarawak,
East Malaysia. Ecology 66: 577-588.
SABAH
FORESTRY
DEPARTMENT
l998 RIL operation guide book.
Sabah Forestry Department, Malaysia.
J-G. 1998. Reduced-Impact logging
SIST,P. and BERTAULT,
experiments: impact of harvesting intensities and logging
techniques on stand damage. In: BERTALLT,
J-G. and KADIR
K.,
(eds.) Silvicultural research in a lowland mixed dipterocarp
forest ofEast Kalimantan, the contribution of STREK project,
CIRAD-Foret Publication pp: 139-162.
SIST,P,, NOLAN,
T., BERTAULT,
J-G. and DYKSTRA,
D. l998 Logging
intensity versus sustainability in Indonesia. For Ecol. Manage.
108: 25 1-260.
THIOLLAY,
J-M. 1992 Influence of selective logging on bird species
diversity in Guiana rain forest. Cons. Biol. 6: 47-63.
UDARBE,
M P , GLAUNER,
R., KLEINE,
M. and U E B E L H ~K.R ,1994
Sustainability criteria for forest management in Sabah. l7TO
Trop. For. Update 4: 13- 17.
UHL,C, and VEIRA,I.C.G. l989 Ecological impact of selective
logging in the Brazilian Amazon : a case study from the
Paragominas region of the state of Para. Biotropica 21: 98106.
VANCLAY,
J.K. 1994 Sustainable timber harvesting: simulation
studies in the tropical rainforests of north Queensland. For
Ecol. Manage. 69: 299-320.
VANUATU
FORESTRY
DEPARTMENT
1997 Vanuatu Reduced-Impact
Logging guidelines. Vanuatu Forestry Department. 23 pp.
WHITE,L.T. 1994 The effects of commercial mechanised selective
logging on a transect in lowland rainforest in the Lop6 Reserve,
Gabon. J. Trop. Ecol. 10: 313-322.
ZAGT,R. J. 1997 Tree demography in the tropical rain forest of
Guyana. Tropenbos, Guyana Series 3, 251 pp.
11
Logging in south Cameroon: current methods and
opportunities for improvement
WYB B.J. JONKERS and GART J.R. VAN LEERSUM
Wageningen University, Sub-department of Forestry, PO. Box 342,
6700 AH Wageningen, The Netherlands
[email protected]. wag-uml
SUMMARY
Forestry operations in the rainforests of south Cameroon require to be adapted to the prevailing physical, biotic and socio-economic
conditions. Terrain used for logging is often steep and rugged. The forest contains many very large trees, though few of those are of
marketable species. The forest is an important resource for the population, and forestry should be planned and executed in close consultation
and cooperation with them. While much effort has been put in improving forest management, logging operations have changed little in
recent years. Less than one tree per ha is felled, and logging damage is therefore limited. About 30% of the felled volume of timber is left
in the forest. Some elements of reduced impact logging, such as winching and improved instruction and supervision, can reduce logging
damage, wastage and negative effects for the local people and wildlife substantially, but other elements, such as liana cutting and directional
felling, are less suitable under the prevailing conditions.
Keywords: Cameroon, forest management, logging, rainforest.
INTRODUCTION
Efforts have been made in many countries to reduce logging
damage in rainforests. Early publications by Mattson MArn
and Jonkers (1981) and Hendrison (1990) indicated that
considerable damage reduction can be achieved by
introducing proper planning procedures and rather simple
modifications in existing logging methods. In the 1980s and
1990s, there was a growing awareness of the need to manage
rainforests in a sustainable way, and this led, among others,
to more attention for reduced impact logging (RIL). Many
studies were executed in Asia, Australia and Latin America
(e.g. Crome etal. 1992, Blate 1997, Johns et al. 1996, Webb
1997, Van der Hout 1999, Bertault and Sist 1995, Pinard and
Putz 1996, Cedergren et al. 1994), and all these advocated
similar changes in logging methods. In Africa, RIL received
less attention. In Cameroon, two RIL studies have been
undertaken. The Tropenbos-Cameroon Programme (TCP)
started in 1994 with RIL and other studies aimed at
developing methods and strategies for sustainable rainforest
management. Another study was carried out in the eastern
part of the country (Durrieu de Madron et al. 1998).
Cameroon has an enormous -diversity of climatic,
physiographic and biotic conditions, and of people.
Rainforests occur in the south of the country. The
200,000 ha Tropenbos research site is located in the South
Province, 50 - 100 km from the coast. Logging methods
should ideally be adapted to the physical and biotic
environment and to socio-economic conditions in the region
for which they are developed. These aspects are therefore
discussed before going into more technical aspects.
THE SOCIAL DIMENSION
The population density in most of the rainforest zone of
Cameroon is 5-15 people km-', but there are also large
uninhabited areas, especially in the east, and densely
populated areas near cities and in areas with fertile volcanic
soils. There are many ethnic groups in the rainforest zone.
In the TCP site alone, there are five Bantu ethnic groups and
the Bagyeli pygmies. The main means of subsistence of the
Bantu is shifting cultivation. Nevertheless, they depend
heavily on bushmeat for their protein supply, and also use
many other non-timber forest products (Van Dijk 1999). The
Bagyeli form a small minority. Although they practise
shifting cultivation, their main means of subsistence are
gathering and hunting. A Bagyeli family is strongly linked
with a particular Bantu clan, with whom they exchange
forest produce for agricultural goods.
Forest management and logging operations should be
compatible with the way of living of the local people, and
with their traditional law. In this traditional law, the concept
of land as property hardly exists. Instead land is regarded
rather as a 'bundle of rights' in respect of which one family
or individual may have the right to practise agriculture on a
12
W.B.J. Jonkers and G.J.R. Van Leersum
particular area, while others may have the right to hunt or to
collect other forest produce. This provides opportunities for
combining timber production and the use of forest resources
by the local people.
TABLE1 Diameter class distribution in a Cameroonian
rainforest
Diameter class,
cm
10-30
30-50
50-70
70-1 00
<l00
Number of stems 379.3
per ha
55.1
17.0
8.5
7.1
PHYSICAL AND BIOTIC FEATURES
The TCP site is located on poor soils of the Pre-Cambrian
Central African Shield. The physiography is undulating in
parts, but more often hilly or mountainous and highly
dissected. Easily accessible parts are mostly used for shifting
cultivation, and logging has, for the most part, to be practised
in difficult terrain. Substantial tracts of land are too steep for
either agriculture or logging, and are used only for gathering
and hunting.
The rainforests of Cameroon vary considerably in species
composition. There is a clear sequence of vegetation types
from the very humid coastal zone towards the somewhat
drier areas further to the north and east. According to
Letouzey (1 985), the rainforest at the TCP site belongs to the
mid-altitude evergreen forest dominated by Caesalpinaceae.
This family is indeed well presented, but the most important
timber species is azobe (Lophiru alata), which is
characteristic of the low altitude evergreen forest. A recent
survey showed a basal area of 34 mZha-', which is well above
the pan-tropical average. The number of species found
among trees >IOcm dbh ranges from 70 to 86 per hectare
(Foahom and Jonkers 1992). Vegetation is greatly influenced
by man. This applies particularly to the extensive areas of
secondary forest, but also to most of the forests considered
'primary'. In past centuries, when the people lived scattered
in the forest, shifting cultivation was practised with primitive
tools, and large trees were often spared on fields because
they were difficult to fell. This seems to have had a long
lasting impact on the forest structure.
A remarkable feature is the high number of very large
emergent trees. Table 1 illustrates this for an inventoried
area of 165 ha. These giants reach heights of 50-60 meters,
with diameters of 1 to 2.5 meters or more. Unfortunately,
only some of these trees belong to marketable species,
and the average logging intensity is well below one tree per
ha. Another noteworthy feature is that the spatial
distributions of large trees of azobC and some other timber
species are clumped. Felling tends to be concentrated in
such clumps, and other parts of the forest are affected little
by felling. As these species regenerate well on shifting
cultivation fields, and hardly in undisturbed forest, such
clumps are believed to originate from such fields abandoned
in past centuries.
The fauna in most forests is not particularly rich because
of hunting pressure. Logging also has an impact on the
fauna. The noise of logging equipment causes larger animals
to move away, and it may take many years before they return.
Moreover. it stimulates hunting as logging tracks provide
easy access to the forest and because demand for bushmeat
increases due to the presence of logging personnel (Van Dijk
1999).
FOREST MANAGEMENT PLANNING
The forestry situation in Cameroon is changing rapidly.
Until recently, gazetted permanent production forest was
almost non-existent, and timber production was in fixedterm concessions of one to five years. Since 1999, shortterm concessions are no longer issued and large permanent
forest management units are being established, for which
forest management plans have to be made. Concepts for
such plans have been developed, and are being elaborated
further. With evolving insights into the complexity of
managing Cameroon's forests, management plans are
becoming increasingly complex. From a purely timber
production oriented plan, the concept has now developed
of a detailed scenario including the rights and obligations
of all actors concerned with the preparation and execution
of the plan (ONADEF 1991, 1997) and local people's
participation.
In Cameroon, forest management planning has to
include land-use planning to determine the exact boundaries
of permanent production and protection forest in
consultation with the local people. After marking the
boundaries, annual logging coupes will have to be planned
in time and space and the annual yield has to be determined,
based on a forest inventory and again in consultation
with the population. To allow people to hunt and to preserve
the fauna, an annual coupe should not cover a large
continuous area and completely surround a village, but
instead consist of logging compartments that are not adjacent
to one another.
Thereafter follows the operational planning of logging
and other forestry activities within the first annual coupe.
A specification of how logging should be executed has to
be part of forest management planning, based on adequate rules and regulations. In Carneroon, however, the
official 'Guidelines for logging enterprises' (MINAGRI
1988) hardly pose restrictions on skidding and felling
other than minimum felling limits, though rules for forest
inventory do apply (ONADEF 1992). A logical set of
harvesting and management guidelines is required for proper
monitoring during, and verification after, the lease period of
a concession. It should be possible to hold a concessionaire
accountable for needless logging damage. Ideally, the
concessionaire should also be obliged to switch to production
technologies which further reduce damage to the residual
stand.
Logging in South Cameroon
CONVENTIONAL LOGGING
TABLE2 Logging damage in twelve 25 ha plots expressed
as % of area disturbed
In order to assess the need for improvements in logging
methods, the operations of one of the best-organised logging
companies in Cameroon and the resulting damage were
studied.
Operation
Logging operations
Damage level
Before logging starts, the company arranges compensation
for possible damage to agricultural fields and other losses
and inconveniences with the villagers. The operations start
with a 100% inventory. Only those trees which the
concessionaire intends to harvest are enumerated and their
positions plotted on 15,000 maps. These trees are always
very large, and produce timber of export quality. Their
average diameter is 116 cm and their average bole volume
13 m3. Maps are used for harvest and marketing planning,
for truck road alignment and for felling, and occasionally
also for skidding.
Felling is done by teams of two or three men. The felling
technique is simple, and trees are usually felled in the
direction of their natural lean. The direction is rarely but
successfully altered in case of possible damage to
agricultural fields. Trees are subsequently crosscut and
topped without any information on log lengths desired
further down the production chain. Felling productivity is
three trees per effective working day per feller.
Skidding is carried out with D7 dozers and Caterpillar
528 skidders. The D7 constructs trails to felled trees and
prepares logs for skidder transport to the landing. Trail
construction is mostly from felling gap to felling gap and is
seldom guided by inventory maps. Logs are skidded one at
a time, and skidding on steeper slopes than 20% is avoided.
The large log sizes and difficult terrain lead to a production
of only five logs per skidder per day. At the landing, logs are
further crosscut to improve their appearance. Log transport
is by trucks with a capacity of 25-35 tonnes.
Logging administration and reporting covers recording
the daily production per crew. The system serves for payment
of bonuses and monitoring of stocks in the forest and on the
landing. The administrative forms for felling, skidding and
transport are however poorly harmonised. As tree and log
numbers on the various forms do not correspond, these
records cannot be used for monitoring the production chain.
Logging damage
Logging damage was studied in twelve 25 hectare plots,
randomly chosen within a 2500 ha working coupe. Only 5%
of the area incurred disturbance as result of logging. Table
2 summarises the results. Damage can be in the form of soil
compaction or vegetation clearing by logging machines, or
damage caused by falling trees. Disturbance is so low
because only 0.3 trees ha-' were felled, which is substantially
less than the 0.7 harvestable trees ha" recorded in the
inventory.
13
Felling Skidding
1.4%
1.1%
Road
Combined
and landing
construction
2.7%
5.1%
The main reason for the low production is that parts of
the forest were not entered because of steep slopes, poor
stocking andlor presence of agricultural fields. It is therefore
not surprising that damage per plot varied considerably,
ranging from 0% to 25% in the plot with the highest felling
intensity (1.8 trees ha-'). Another reason for the low yield
was that the concessionaire had temporarily increased the
minimum felling diameter and reduced the list of species to
be harvested because of low timber prices.
It is remarkable that half of the damage was caused by
truck road and landing construction, which is usually a
minor cause. This is because the terrain is highly dissected,
and roads were built on each of the many ridges.
Furthermore, roads were made very wide to allow quick
drying after rains. Log landings were also oversized, and
unnecessary landings were made. The crews had been
instructed to create landings every 500 meters, which was
observed rigidly. About 30 % of the damage due to roads and
landings could have been avoided.
Avoidable skidding damage includes dual trails, shortcuts
and trails not leading to a felled tree. About 20% of this type
of damage could have been avoided, and was caused by:
-
Rain and unstable, saturated ground conditions. This is
considered the single most important factor. If skidding
continues in bad weather the output is almost zero,
leading to rapid deterioration of the trails, which can
sometimes be used for only one passage.
- Lack of supervision. Machine operators can set their
own standards of work. As long as they produce enough
logs, their supervisors remain on the landing and inspect
only logs that reach there.
- Lack of environmental awareness among operators and
supervisors. Few operators see a need for damage
reduction, and most feel that the forest will recover
anyway.
- Dozers entering felling gaps. Needless manoeuvring in
felling gaps caused 10% of the total skidding damage.
Reducing the area of felling gaps by making gaps overlap
is not feasible, as distances between felled trees are generally
too large. The large distances between harvestable trees also
reduce the need for careful planning of felling directions to
facilitate skidding, as it is always possible to approach logs
using the best possible angle. Directional felling may be
useful for preserving future crop trees and trees producing
non-timber forest products.
14
W.B.J. Jonkers and G.J. R. Van Leersum
Timber recovery
Summary results
Reducing timber wastage is an aim of RIL, as it makes the
operation economically more attractive. Timber losses were
therefore investigated. It was found that the quantity of
timber delivered at the sawmill was 70% of the amount
felled. Most losses occurred during felling (21 7%). As much
as 7-10% was lost because good-quality tops of the stems
were cut off and left in the forest. Other felling losses
consisted mostly of inferior timber such as conical butt ends
and hollow or poorly shaped trunks. Losses due to skidding
(4%) consisted mainly of good-quality logs left along trails.
At landings, another 4% loss occurred, mainly due to cutting
off both log ends to give the timber a better appearance.
Some good logs were left at landings due to oversight.
Underlying causes for timber losses during logging
are:
Liana cutting
- High quality demands imposed by the sawmill and sales
branch of the company.
- Natural factors such as tree rot.
- Poor reporting and administrative procedures. The log
-
-
flow is not monitored in a way that allows timber losses
to be traced.
Inadequate monitoring and supervision, especially in
relation to the quality of the work, e.g. assessing abuse
of machines, need for training among operators, wastage
and logging damage.
Method of remuneration and incentives for forest
workers. These payments are based on quality and
quantity of timber arriving at landings and not on easily
obtainable indicators for the quality of their work such as
the volume recovered per tree harvested.
REDUCED IMPACT LOGGING
Conceptual framework
When devising a RIL method, a wide range of harvesting
operations and pre- and post-logging activities has to be
considered. Box 1 provides a list of such operations, which
is based on publications mentioned above (p. l ] ) , plus
others (e.g. FAO 1996) as well as the authors' views at the
onset of the project. This list was used as a conceptual
framework in TCP logging research. The study looked into
the technical feasibility of the elements and the need for
their introduction in Cameroon.
Field experiments
Potential adaptations to conventional logging aimed at
damage reduction were identified and tested individually.
Thereafter, a field comparison (120 ha) was made between
conventional and modified logging: analysis of this
experiment remains to be completed. In addition, the impact
of liana cutting on logging damage was assessed in a 28 ha
experiment. Liana cutting nine months before felling was
compared to a control treatment.
Liana cutting did not have a noticeable effect on felling
damage, in spite of the large numbers of lianas present
(Parren and Bongers 1998). This is probably because felled
trees are all emergents which tower high above surrounding
trees. Lianas in such emergent trees seldom connect the
crowns of these trees to neighbouring crowns, and therefore
contribute little to felling damage.
Harvest inventory and skidding trail alignment
With the available level of mapped topographic detail, tree
inventory data are not sufficient to plan logging operations
in difficult, irregularly dissected terrain. Enlarged
topographic maps used to plot inventory results are too
inaccurate for 'precision forestry' purposes. When designing
a system of skidding trails, a double field check is needed
before marking trails. Even then, frequently used trails
deteriorate, forcing the machines to deviate from the
intended pattern. Alignment of trails based on detailed
inventory maps was tried. The outcome was that the skidding
trail pattern is determined mainly by terrain conditions.
However, involving local villagers in trail alignment with
the aim of protecting sites important to them is a promising
option, as is the use of topographic information from aerial
photographs, remote sensing or radar imagery. Geographic
Information Systems can also be of use (Durrieu de Madron
et al. 1998).
Felling
Directional felling proved to be technically feasible. Only
the very few trees without aclear natural lean are difficult to
fell in a desirable direction. Directional felling can serve to
protect potential crop trees and trees yielding non-timber
forest products. However, damage to trees to be preserved
is low under conventional felling and preliminary results
indicate that directional felling hardly reduces damage. The
main reasons are the scattered distribution of trees to be
preserved and poor horizontal visibility, which is usually 30
metres or less while the bulk of a felled tree's crown
penetrates the crown layer to a greater distance. This makes
it difficult for fellers to choose a proper felling direction. By
far the biggest improvements are expected from refresher
courses in controlled felling and improved crosscutting
instructions.
Skidding
Skidding offers little scope for improvement along lines
developed outside Africa. Reduction of tertiary skidding
trails through winching and pre-felling alignment of trails
on the basis of a detailed harvest inventory were considered promising techniques in this respect, and have
been tried. Winching over long distances was seldom
Logging in South Cameroon
possible because o f log weights a n d volumes, obstruction
b y t h e bucked, conical butt e n d a n d hilly and slippery
terrain. Winching o v e r short distances gave better results.
Supervision of skidding operations has immediate posi-
tive effects o n d a m a g e reduction. U s e restriction following rain also reduces d a m a g e , but leads t o losses o f output
in all production phases w h i c h m a y o u t w e i g h t h e
benefits.
Box 1 Potential elements for a RIL method for tropical rainforest
Harvesting phase:
Featureslissueslpurpose
Inventory
Scale of maps
Full topography
Inventory unit size
Markinglmapping of timber trees
Markinglmapping of PCTs
Markinglmapping of NTFP trees and sites
Report
l : 5000, balance between information density and size
Location of breaks in terrain,obstacles to skidding
20-50 ha, stock taker density and communication
Including their natural lean
Preserve potential crop trees (PCTs)
Safeguard supply of non-timber forest products (NTFPs)
Submit information to planning department
Planning
Assignment of buffer zones
Tree selection:
Assignment of trees to be preserved
Assignment of trees to be felled
Felling pattern
Felling direction
15
Forest protection within working coupes
Skidding trail alignment
Report
Report
Seed trees, trees of importance to local people
Percentage of harvestable stock
Scattered or clumped
Avoiding NTFP trees, PCTs, adverse angle with trail, single
or multiple tree felling gaps
Design system of unambiguous, shortest tracks
Relay planning outcome to local people1 prescriptions to field crews
Feed back alignment results to planning
Other pre-felling activities
Climber cutting
Girdling of timber trees
Marking of skidding trails
Well in advance, at individual tree level
Reduction of crown weight
Extra orientation for felling direction of trees
Road a n d landing construction
Clearing width
Wildlife corridor
Durability
Depending on orientation towards sun
Connected crowns of bordering trees
Shape, compactness, top layer material
Felling
Tree selection
Felling direction
Directional felling
Cross-cutting, incl. topping
Report
Only felling of trees indicated by planning
Check whether planned direction is realistic
Avoid damage to PCT and NTFP trees, avoid unfavourable angles
Avoid skidding problems, according to log size preferences in market
Relay information to skidding crews, feedback to planning department.
Skidding
Marking of skidding trails
Construction of skidding trails
Construction of stream crossings
Stump operationslwinching of logs to trail
Skidding to landing
Report
Present unambiguous trail to operator
Computer aided, possibly computer controlled
Hollow trees as culverts, feller present
Reduction in tertiary trailslstump site damage, communication
Remain on trail, terrain conditions
Relay information to transport crew, feed-back planning department
Post-harvest operations
Felling gaps
Skidding trails
Main roads
Feeder roads
Bridges
Landings
Sanitary operations on trees, diseases, growth stimulus
Ploughing, blocking, removal of culverts
Keep open for local populat~on
Blocked to prevent re-logging and hunting
Blocked to prevent re-logging and hunting
Ploughing, 60 cm depth
16
W.B.J. Joizkers and G.J.R. Van Leersum
DISCUSSION
Logging damage has been measured in the concession of a
comparatively well organised enterprise which applies
planning a n d control. Although logging methods have
improved little in recent years (see Evans 1990, Gartlan
1990), comparison of this operation with the full range of
R I L elements is unlikely t o show dramatic differences under
the current low harvest intensities. About 20% of the damage
can be avoided, meaning that logging damage could have
been 4% instead of 5%. N o doubt a comparison with
practices of less organised enterprises puts R I L in a more
favourable position.
Nevertheless, there is a need t o reduce damage not only
t o preserve future crops, but also t o reduce negative effects
for the local population and wildlife. The concepts and
activities t o achieve this are not new, with the exception of
involving the local population directly in forestry activities.
Better communication, reporting and supervision will
diminish damage and improve the recovery of felled timber.
ACKNOWLEDGEMENTS
T h e International Tropical T i m b e r Organisation, t h e
C o m m o n F u n d f o r C o m m o d i t i e s and t h e Tropenbos
Foundation funded the research on which this paper is
based. T h e authors are grateful for the support of WijmaDouala S A R L and all those w h o contributed to this study, in
particular Messrs F. Ngibaot and E . Laan.
REFERENCES
BERTAULT,
J-G and SIST,P. 1995 Impact de l'exploitation en foret
naturelle. Bois et Forets des Tropiques 245 ( 3 ) : 15-20.
BLATE,
G. 1997 Sustainable forest management in Brazil. Tropical
Forestry Update 7 (3): 14-15.
CEDERGREN,
J., FALCK,
J., GARCIA,
A., GOH,F., and HAGNER,
M.
1994 Reducing impact without reducing yield. Tropical
Forestry Update 4 (3): 9- 10.
L.A. and RICHARDS,
G.C. 1992 A study of
CROME,
F H J, MOORE,
logging damage in upland rainforest in north Queensland. For.
Ecol. Manage. 49: 1-29.
DE MADRON,
L., FORNI,E. and MEKOK,M. l998 Les
DURRIEU
techniques d'exploitation 8 faible impact en foret dense humide
camerounaise. SCrie Forafri Document 17, CIRAD-Foret,
Montpellier, France.
EVANS,
W.R. l990 The sustainability of logging in Cameroon:
selected case studies. Fountain Renewable Resources,
Banbury, United Kingdom.
FAO. 1996 Forest Codes of Practice: contributing to
environmentally sound forest practices, FAO Document 133.
FAO, Rome, Italy.
FOAHOM,
B. and JONKERS,
W.B.J. 1992 A programme for Tropenbos
research in Cameroon. The Tropenbos Foundation,
Wageningen, The Netherlands.
GARTLAN,
S. l990 Practical constraints on sustainable logging in
Cameroon. Paper presented at the ConfCrence sur la
conservation et I'utilisation rationnelle de la foret dense
d' Afrique Centrale et de 1'Ouest. African Development Bank,
IUCN, World Bank, Government of CBte d'lvoire.
HENDRISON,
J. l990 Damage-controlled logging in tropical
rainforest in Suriname. PhD thesis, Wageningen Agricultural
University, Wageningen, The Netherlands.
JOHNS,
J. S., BARRETO,
P. and UHL,C. 1996 Logging damage during
planned and unplanned logging operations in the eastern
Amazon. For. Ecol. Manage. 89 ( 1 ) : 59-78.
LETOUZEY.
R. 1985 Notice de la carte phytogCographique du
Cameroun. P. Lechevalier, Paris, France.
MATTSON
MARS, H, and JONKERS,
W.B.J. 1981 Logging damage
in tropical high forest. Project FAO/MAU76/008 working
paper 5, Sarawak Forestry Department, Kuching, Malaysia.
MINAGRI. 1988 Guidelines for logging enterprises. Ministry of
Agriculture, YaoundC, Cameroon.
ONADEF. 1991 Canevas de Plan d'AmCnagement Forestier.
Office National de DCveloppement des ForCts, Yaoundt,
Cameroon.
ONADEF. 1992 Normes d'inventaire de reconnaissance des
ressources forestihres. Office National de DCveloppement des
Forets, YaoundC, Cameroon.
ONADEF. 1997 Guide d'ilaboration des plans d'amknagement
des forets de production du domaine forestier permanent de la
Rtpublique du Cameroun. Office National de Dtveloppement
des ForCts, YaoundC, Cameroon.
PARREN,
M. and BONGERS,
F. 1998 Forest lianas and pre-felling
climber cutting in southern Cameroon: a silvicultural
evaluation. Paper presented at the Forafri conference,
Libreville, Gabon, October 1998.
PINARD,
M.A. and PUTZ,F E. 1996 Retaining forest biomass by
reducing logging damage. Biotropica 28 (3):278-295.
VANDER HOUT,P. l999 Reduced impact logging in the tropical
rainforest of Guyana. Tropenbos-Guyana Series 6. TropenbosGuyana Programme, Georgetown, Guyana.
VANDIJK.J.F.W. l999 Non-timber forest products in the BipindiAkom I1 region, Cameroon. Tropenbos-Cameroon Series 1.
Tropenbos-Cameroon Programme, Kribi, Cameroon.
WEBB.E L. 1997 Canopy removal and residual stand damage
during controlled selective logging in lowland swamp forest of
north-east Costa Rica. For Ecol. Manage. 95: 117-129.
I
17
RIL for real: introducing reduced impact logging techniques
into a commerical forestry operation in Guyana
S. ARMSTRONG and C. J, INGLIS
Edinburgh Centre for Tropical Forests, c10 LTS International, Pentlands Science Park, Bush Loan, Penicuik,
Edinburgh, EH26 OPH, U.K.
SUMMARY
The commercial implications of introducing Reduced Impact Logging (RIL) into a large timber harvesting operation are assessed. Findings
are based on a trial of 800 hectares in a timber harvesting operation in Guyana, and indicate that damage can be reduced and financial
savings made by implementing improved harvesting practice, particularly through the use of pre-harvest tree location maps. Adequate
training and supervision are central to implementing good harvesting practice. Major constraints to implementing RIL are the availability
and retention of people with the necessary skills to work in difficult conditions. Adoption of new technologies and improved information
and management systems should also improve operational efficiency and reduce damage in other areas of commercial forestry operations
such as road construction.
Keywords: commercial forestry harvesting, forest management, Guyana, reduced impact logging.
INTRODUCTION
The problems of implementing good forest management are
not new. In drawing lessons from the history of tropical
forest management Dawkins and Philip (1998) highlight the
importance of the effective control of felling in order to
avoid both waste of and damage to timber and the residual
stand. They cite Maconachie and Tremenhere as having
made this same point in 1810 and 1841 respectively.
Actually achieving these aims, however, still proves problematic in commercial timber harvesting operations.
The research trial presented here illustrates some of the
practical implications of adopting RIL theory and techniques into a commercial operation. Such a study has been
called for by an industry wary of introducing RIL (see, for
example, ITTO 1996).
APPROACH
Since 1992, the Edinburgh Centre for Tropical Forests
(ECTF) has been working together with the Barama Company Limited (BCL), a commercial timber harvesting and
processing company, on the North West Guyana Sustainable
Timber Production Programme. The aims of the Programme
partners are:
BCL: to ensure 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;
ECTF: to provide independent advice to BCL, to enable
it to fulfil its wider management objectives through a
process of environmental and social monitoring and a
programme of silvicultural and operational research.
Earlier work undertaken by the programme, notably the
establishment of Permanent Sample Plots (PSPs), is described in more detail elsewhere (Inglis et a1.1997). These
plots are being used to assess post-harvest forest growth and
the impact of damage on growth. The background to Guyana
and forestry in Guyana are described in Inglis et al. (1997)
and van der Hout (1999).
SITE DESCRIPTION
BCL operates a 1.67 million hectare concession in North
West Guyana. The forest in which the trial was undertaken
is a mixed evergreen moist forest on undulating to hilly
terrain. The average basal area recorded in trees above
20 cm dbh in 78 one hectare PSPs established in the
concession is 23 m ' ha-', ranging between 15 and 27 m2
ha-'. The productive mixed forest comprises mainly black
kakaralli (Eschweilera spp.), kauta (Licania guianensis),
hairiballi (Alexa spp.), trysil (Pentaclethra odorata) and
baromalli (Catostemma spp.), representing 23%, 14%,
11 %, 6% and 6% respectively of the basal area in the PSPs.
Of these baromalli and trysil are harvested. Stocking of
commercial species is low, with harvesting averaging
18
S. Armstrong and C.J. Inglis
8.2 m3 ha-' since operations began. At an average volume of
3.4 m3per tree, this implies 2.4 trees per hectare. Baromalli
comprises nearly 90% of the harvested volume. The
productive forest is also interspersed with unproductive
swamp and steep terrain.
The harvesting in this trial was subject to the many
interacting and uncontrolled factors common to tropical
forest harvesting operations such as variable ground conditions and machine age. For the results of the trial to feed into
management planning it was important that the results would
be 'operationally relevant'. As such, operational variables
(especially scale) were retained at the expense of conducting
a less realistic but more precisely controlled experiment, and
the results focus on the implications of implementing RIL at
the operational scale. A smaller scale and more intensive
study of the ecological and silvicultural impacts of RIL
elsewhere in Guyana is described by van der Hout (1999).
Definitions
For the purposes of this study the following terms are
defined.
Skid trails1
The proposed main skid trail length is the length of the main
skid trail as planned on the pre-harvest tree location map. It
does not include the secondary skid trails, which run from
the main skid trail to the stumps. In practice skid trail
planning is subjective and variation between experienced
individuals planning skid trails is likely.
Waste wood
Waste wood is defined here as portions of felled logs over
1 metre in length which are merchantable but which have not
been extracted. This is usually because the log has been cut
short or missed by the skidder and left behind. It does not
include stumps or wood that is trimmed after extraction nor
volume of wood lost to poor harvesting techniques, e.g.
through splitting.
Skid trails are marked using flagging tape prior to harvesting. While the bulldozer clears the skid trails the feller walks
the block looking for and felling merchantable trees. After
felling the bulldozer positions the log at stump to ease
extraction by the skidder. The skidder then pulls the log
from the stump along the skid trail to the roadside. There is
little winching of the log during the process. This system is
taken to be 'conventional practice'.
DESCRIPTION OF THE LOGGING TRIAL
The work described here developed from studies of wood
waste, a matter of concern as it was perceived to be
restricting operational efficiency. These waste wood studies
revealed additional areas for improvement, specifically skid
trail layout, and these in turn focused research on operational organisation and machine use. The focus and
methodology of the work has therefore evolved as new areas
for investigation emerged. As the study progressed additional techniques intended to improve harvesting practice
were adopted, moving towards what is now termed reduced
impact logging. Definition of RIL should at least imply a
systematic approach to harvesting, specifically improved
pre-harvest planning on the basis of appropriate and accurate information. The first blocks to be harvested, numbered
1 and 2 in Table 1, may be thought of as conventionally
logged and each subsequent block to more closely approach
this definition of RIL. The trial began in 1996 with block
1 and ended in 1999 with block 8. The composition of the
harvesting team changed over the course of the trial as
operators moved out of the area to return to their families or
to take up other work.
T A B L E1 Treatments of blocks in the trial
Key to column numbers:
1 : block
2: trails marked out based on map
3: pre-harvest map given to team
4: increased supervision
5: vine cutting
6: directional felling
training
Uncut merchantable timber
Uncut merchantable timber is defined here as merchantable
and accessible trees that have not been felled. It is assumed
that these trees were missed by the feller.
CURRENT LOGGING PRACTICE
BCL uses a 'blocking' system, where the forest is divided
into 100 hectare blocks (lkm square) which are harvested
sequentially. Eight of these blocks form the basis of this
trial. The boundary of each 100 hectare block is marked,
and a 10% pre-harvest inventory conducted using four 25m
by lkm strips within the block. A selective harvesting
system is practised using bulldozers and wheeled skidders.
A 100% inventory of all potentially merchantable trees
was completed in each block. The inventory team walked
the block along 50m wide strips divided into forty 25m long
l
The usage 'skid trail', as opposed to 'skidding trail ' is retained
throughout the paper. Ed.
RIL for real
stations in each strip. Each tree was individually numbered
with a tag. During the inventory of the last 6 blocks the
location of enumerated trees within each strip and station
was recorded and this was used to produce a pre-harvest tree
location map at a scale of 1:2,500. The location of features
restricting harvesting, e . g . steep slopes, streams, was also
marked. This map was used to plan the layout of the main
skid trails and to locate trees for harvesting.
In the last 5 blocks the operation of the machinery was
timed over the whole harvesting period and recorded by
machine activity. After harvesting the length of skid trails
was surveyed. In most blocks the number of merchantable
trees that were not felled, merchantable logs felled but not
skidded and the volume wasted from logs being cut short
were recorded.
Description of harvesting in each block
Blocks 1 and 2 were logged using 'conventional practice'.
The pre-harvest tree location map was not used to mark out
skid trails and harvesting teams were not given any prior
indication of the position of merchantable trees.
In blocks 3 and 4 main skid trails were marked out using
the pre-harvest tree location maps. In block 3 the preharvest tree location map was not given to the harvesting
team prior to harvesting, whereas in block 4 the team was
given the pre-harvest tree location map. Blocks 3 and 4
have relatively difficult terrain for harvesting with intersecting steep slopes and swamps.
In blocks 5 and 6 skid trails were marked out and the
pre-harvest tree location map was given to harvesting
teams before harvesting. In addition, the harvesting team
was closely supervised, and had gained experience of
using pre-harvest maps whilst harvesting blocks block 3
and 4. The blocks are both flat with significant areas of
swamp.
In blocks 7 and 8 vine cutting was practised in the blocks
during the 100% inventory (about 6 months before harvesting). Trees were felled using wedges in both blocks.
Professional training in chainsaw use and directional felling
techniques was given after felling block 7 and prior to
starting work in block 8. In both blocks tree inventory
numbers were recorded on all logs and checked at roadside
against the pre-harvest 100% inventory to ensure all accessible and merchantable trees which had been inventoried
were extracted. As trees were felled and skidded they were
'ticked off' on the tree location map. The harvesting team
was well motivated, experienced in the techniques used,
closely monitored, and used to working alongside the
research team. Waste wood was not measured in these
blocks.
RESULTS
Harvesting results are summarised in Table 2. Figure 1
shows the layout of the proposed main skid trails and actual
total skid trails in blocks 3 and 5.
TABLE2 Summary results
-
Block
Skid trail Total actual length, km 13.2
Proposed main length, km
n/a
Total actuaVproposed main
nla
Harvest Volume m3
Number of logs
Uncut merchantable trees, no.
Proportion of harvest
Waste wooda, m'
No, pieces
Proportion of harvest
Left logs, m3
No. logs
% of harvest
1,953
516
22
4%
38
60
2%
5
1
1%
13.2
5.4
2.46
2,255
636
n/a
nla
110
63
5%
58
16
2%
Bulldozer operating time (hrs:min)
Total recorded time
Positioning logs for extraction
Skid trail construction & overhaul
Position logs for extraction
(per 100 logs harvested)
Construct and overhaul
(per km skid trail constructed)
Note: a, waste wood volume includes that of left logs
2,299
627
22
3%
38
19
1%
29
6
1%
19
2,024
564
0
0%
n/a
n/a
nla
0
0
0%
227:32
34:20
3152
06:05
02:24
20
S.Annstrong and C.J. Inglis
Block 3
Block 5
Proposed
Actual
FIGURE
l Skid trailpatterns
Skid trails
As shown in Table 2, there was a significant difference
between the length of the proposed main skid trail and the
actual total skid trail length in each block.
Figure 1 indicates a tendency to produce an inefficient
skid trail design where no pre-harvest tree location map is
used, as in block 3. Deviation between proposed skid trail
layout and actual layout is still apparent in block 5, even
though the team was given the pre-harvest block map. The
ratio of actual total skid trail length to proposed main skid
trail length is over 200% for all but the last block in the
trial.
Waste wood
The total production volume lost as waste wood was around
3% but decreased as the trial progressed. The volume lost
in high stumps is not recorded here, but in other studies was
found to be consistently very low (less than 0.01 m' ha-' ).
The volume of timber lost to poor felling technique, e.g.
splitting or drawn fibers, is not recorded, but observation
suggests that experienced and well-trained fellers can
significantly reduce the incidence of wood lost as a result of
poor felling technique.
Around 20 trees were missed by harvesting teams in each
block, except in the last two, where no trees were missed.
Analysis by species in the first six blocks combined
revealed that 89% of the inventoried baromalli were
harvested, compared with 69% for other merchantable
species.
Time analysis
Machine working time for the bulldozer is given in Table 2.
Total working time in each block was very variable,
indicating the importance of uncontrolled operational
factors in harvesting.
Observation revealed that directional felling (block 8) of
a tree took longer than conventional felling (around 7
minutes against 3 minutes for trees of about 60 cm diameter). The increased time to fell trees was not a constraint
on production but was more tiring for the feller. In this block
the bulldozer time required for positioning logs was significantly reduced.
Basis for cost analysis
This trial provides indication of the additional resource
requirements and benefits that may be expected from
implementing RIL techniques. While actual recorded costs
cannot be quoted, and the costs of implementing RIL
techniques will vary between operations, the basis for
calculation presented here may provide a useful guide for
analysis of other operations.
RIL for real
The costs of implementing RIL are associated with
increased planning and more developed management and
information systems. The financial benefits arise specifically from increased production with less waste and decreased cost from the more efficient use of machinery.
In assessing the cost implications of adopting RIL
techniques the following elements were considered.
21
Supervision and training
Allowance was made for increased supervision and training.
Estimation of these costs per unit output is fairly subjective,
but the absolute costs are small in relation to the value of
operating costs and outputs.
Outputs
Pre-harvest tree location map
Dependent on terrain and stocking density an enumeration
team of 5 men can cut strip lines at 100 m intervals in a block
in about 5 days, enumerate the block in 3 days, and mark the
skid trails in 2 days; a total of 10 days. With experience,
teams will complete blocks faster. Reducing the minimum
diameter inventoried can significantly increase the number
of trees to be inventoried and hence the time the inventory
takes. Trees were recorded to 50 cm dbh in this trial. The
enumeration team needs to be proficient in inventory
techniques, notably tree spotting and assessment of
merchanability. They should be sufficiently well motivated
to work without continuous supervision and be able to
record data for map making.
A draughtsman can produce the pre-harvest tree location
map for a 100 hectare block map in a day.
Vine cutting
In this trial cutting all vines in the block was found to be
excessively time-consuming. It was only feasible to cut
vines around those trees identified for harvesting and one
man could clear the vines around 40 trees in a day.
Data processing
Enumeration data entry to enable monitoring of harvested
logs against the inventory and for production forecasting
requires the skills of a data entry clerk for around one day
per block.
Making skid trails
The time taken to build skid trails and their cost are heavily
dependent on terrain and ground conditions. On the basis of
results from the trial 2 hours of bulldozer running time per
km of skid trail constructed was used for costing. Efficiency
of skid trail layout and construction also depends on the
level of experience, training and supervision available.
Based on the trial a 30% reduction in skid trail length seems
possible by providing a tree location map, basic training and
increased supervision.
Based on the results of this trial an increase in production of
7% per block was projected. This figure was based on an
expected reduction in wastage as the proportion of uncut
merchantable standing timber and missed felled logs is
reduced and recovery from cross cutting logs is improved.
There should also be increased output as fewer logs are
damaged during felling by poor felling practice.
Results of cost analysis
The saving in machine costs (mainly bulldozer time for skid
trail construction) possible by adopting the RIL techniques
described here marginally outweighed the resulting additional costs. Varying the intensity and extent of inputs, e.g.
the number of man days spent enumerating or supervising,
changed the absolute value of the saving. However, when
the value of the increase in production per block (because of
reduced waste) was added into the calculation it was found
that the adoption of RIL techniques significantly reduced the
production costs per m3. This finding is in line with other
studies reviewed for example in Boltz, Holmes and Carter
(1998), CURE (1998) and van der Hout (1999).
Savings are also made at the level of the company's
operations. With increased production per block there is a
reduction in the area of forest needed per m' harvested. This
implies an additional cost reduction in the relatively expensive activities of road construction, road maintenance and
hauling per m3 produced.
At the operational scale, however, additional costs may
be associated with undertaking the pre-harvest enumeration.
It takes a team up to 10 days to undertake a 100% inventory
of 100 hectares and produce a pre-harvest tree location
map. To inventory at the same rate as harvesting it is likely
that additional staff will need to be employed and trained.
For each of these people there is an additional requirement
for transport, supervision and administrative support. Availability of additional staff and the cost of these additional
overheads could be significant constraints to adopting
RIL.
DISCUSSION
Positioning the log for extraction
Skid trail layout
Directional felling reduces the time taken to position the log
at the stump for extraction. A reduction from 8 to 2.5
bulldozer working hours per 100 logs harvested was used in
the calculation on the basis of the results of the trial.
Swampy areas, heavy clays and rain in the operating area
make skid trail construction and skidding difficult. Part of
the difference between the actual total skid trail length and
the office based proposed main skid trail is accounted for
22
S. A m s t r o n g and C.J. inglis
in the length of secondary skid trails which are not included
in the proposed main skid trail length. Differences also
resulted when the proposed skid trail layout was found to be
impractical on the ground, usually because of recent heavy
rainfall making ground conditions impassable. The length
of skid trail, proposed or actual, is strongly determined by
the topography of the block, ground conditions and distribution of merchantable trees.
Office-based skid trail planning is only effective with
extensive ground checking and requires a good understanding of the practical operating constraints of harvesting
machinery. For harvesting teams to keep to proposed skid
trails, training for machinery operators in map interpretation
and good supervision are important.
Observation revealed that the most destructive damage
in the forest resulted from the action of the bulldozer, both
in building skid trails and in positioning the log for
extraction. Controlling the activity of the bulldozer, as by
decreasing the length of skid trail built, reduces the level of
damage in the remaining stand. Damage caused by clearing
trees with attached vines during skid trail construction was
also significant. Vines should be mapped during enumeration to avoid the damaging practice of building skid trails
through areas of high vine density.
Efficient use of machinery implies that machines are
used to extract timber without engaging in other, nonessential activities. The most efficient operations would
undertake clearance only of trees which were essential for
timber extraction. Causing unnecessary damage to the
residual stand is costly since it involves bulldozer or skidder
time for no financial gain. As such, an efficient operation
should optimise the use of machinery and minimise the
damage to the residual forest.
interpreting the tree location map and able to direct the feller
to harvestable trees. By checking the tags of the trees
extracted to roadside against the inventory results in these
last 2 blocks, it was also possible to locate unharvested
merchantable trees readily using the pre-harvest tree location map, and ensure that all potential trees were extracted.
Several trees that might have been missed were harvested as
a result of this system.
Baromalli is the dominant merchantable species and the
results indicate that it is less likely to be missed than other
species. One reason for this may be that fellers are more
proficient at identifying the most common commercial
species than other species. Training fellers in spotting other
commercial species should reduce the production volume
lost in this way.
Waste wood
Management systems
Fellers cut logs shorter than the full merchantable length.
Fellers often cut short of the first branch or defect because
of the difficulties in accessing the top end of the log caused
by the tangle of crown debris brought down by felling. The
trial nevertheless suggests that improved supervision and
training effectively reduce the level of waste wood. Where
whole length trees were not extracted this was because of
poor co-ordination between skidder driver and feller. Information on tree and log locations is readily available in the
block map, and needs to be conveyed to the skidder operator.
Pre-harvest tree location maps are effective only if there is
also robust communication between the operators in the
harvesting team.
Information and monitoring systems need to be robust and
simple if they are to succeed in the often unpredictable
environment of a tropical timber harvesting operation. It is
particularly important to remember the objectives of the
harvesting system, notably to reduce damage and improve,
efficiency, rather than perceiving the objective to be production of tree location maps per se.
Improvements do not need to be all or nothing to have an
impact. Incremental improvements in planning and supervision can improve efficiency and reduce damage. Subjective post-harvest monitoring of wood waste and skid trail
length which feeds back into management practice, for
example, can be effective and readily undertaken with
limited resources.
Assessing the viability of RIL in other harvesting operations
The impact of improved harvesting techniques will vary
between different commercial operations. An instructive
exercise for any commercial operation would be to assess
the volume of wood left unharvested in a recently closed
harvesting area and also the extent and layout of skid trails.
A quick comparison of the relative costs of machinery, the
loss of production and the costs of an inventory crew may
indicate the financial benefits of implementing RIL techniques, or at least of strengthening the planning process. To
sustain such changes over the long term the operation should
also assess its capacity to handle the anticipated increase in
flow of information effectively, to support additional inventory staff, to strengthen its field supervision and to provide
training.
Training a n d skills
Possession of a tree harvesting map does not itself reduce the
incidence of potentially merchantable trees that are missed
by the fellers, other factors are important. In the last two
blocks, blocks 7 and 8, where no merchantable trees were
missed, Ehe feiler's helper was particularly proficient in
Acquiring technical skills, such as directional fellin
winching, is implied in the adoption of RIL. In addition,
checking that trees recorded in the pre-harvest inventory
have been extracted to roadside or reading a tree location
*"
RIL for real
map requires data handling and data analysis skills. Such
paper-based skills have not been commonly required in
harvesting teams up to now and requiring them may imply
employment of people with a more comprehensive range of
literacy skills than at present.
For RIL to be sustainably implemented trained staff need
to be retained. Keeping skilled staff in the dangerous,
physically demanding environment of a harvesting team is a
major constraint to implementing RIL techniques. This is
especially true as skilled workers often have increased
opportunities for obtaining work elsewhere. It seems likely
that offering incentives, whether financial or through improved working conditions, will be important in sustaining
RIL commercially.
23
CONCLUSIONS
Pre-harvest tree location maps are a key part of improving
harvesting but they also need to be used in combination with
adequate supervision and training of operators. Information
gained in the pre-planning phase should be fed into a
management system that includes monitoring and control.
Improved efficiency of skid trail construction implies
reduced damage and reduced cost at the operational scale.
The main constraint to implementing RIL in an operation are
the availability and retention of people with the necessary
skills in difficult working environments.
Developing strong management and information systems is therefore a key part of implementing RIL in
commercial operations.
Wider operational considerations
Whilst this study has focused on harvesting it seems likely
that improved planning and supervision should also improve
efficiency and reduce the damage caused by other parts of
the operation such as road construction. Actual financial
gain will depend on the cost of additional planning and
information relative to improvements in efficiency such as
a reduction in road density. Improved planning and stronger
information and management systems can be hard to
implement in institutions especially where the necessary
human resources are lacking. Larger operations may be
better placed to support the necessary institutional changes
as the required additional overheads may be spread over
a higher output. Similarly Souza (1999) suggests a
minimum operational size in determining the financial
feasibility of introduciAg GIS facilities into tropical timber
operations.
By raising skill levels, adopting new technologies such
as hand-held data loggers and computerising data systems
the planning process could be speeded up and made more
accurate. Tropical forest harvesting, like any other industry,
needs to embrace new technology and develop its skills base
if it is to optimise the use of the available resource.
Barama after the trials
Within the Company the trial has served to highlight
particular areas for improvement in harvesting practice and
efficiency. The importance of close supervision is seen as
key to the process and the role of 'block inspectors', who
plan skid trails and monitor blocks post-harvest, has been
strengthened. An additional harvesting supervisor has also
been employed. Pre-harvest block mapping is being undertaken at the pilot level to determine the practicality of
implementing 100% enumeration and the production of preharvest tree location maps throughout the operation. Specific training programmes are now being developed for
operators, inventory teams, surveyors and block inspectors.
The company is also exploring the option of changing the
harvesting system so that on flat ground the bulldozer will
only construct the main skid trail.
ACKNOWLEDGEMENTS
This study has been made possible by the Barama Company
Limited. The field work was completed by the BCL research
team in conjunction with ECTF. Greg Sutton, Andrew
Leslie and Mark Lawrence each made an invaluable contribution to this process. Thanks are due to all.
REFERENCES
BOLTZ,
F., HOMES,
T.P. and CARTER,
D.R. 1998 The economics of
reduced Impact logging in the American Tropics: A review of
recent initiatives. Florida Agriculture Experiment Station
Journal.
DAWKINS,
H.C. and PHILIP,M.S. 1998 Tropical moist forest
silviculture and management: A history ofsuccess andfailure.
CAB lnternational, Wallingford, Oxon.
CURE. 1998 Low impact logging study shows that LIL increases
efficiency, reduces ecological impact and saves money!
C. U.R.E News. lnternational Wood Products Association
Conservation Utilization Reforestation and Education
Programme.
INGLIS,
C.J., SUTTON,
G. and LAWSON,
G.J. 1996 Research and
monitoring for sustainable forest management in NW Guyana.
In: DYKSTRA,
D.P. and HEINRICH,
R. (eds.) Research on
environmentally sound forest practices to sustain tropical
forests. FAOIIUFRO Satellite meeting, IUFRO XX World
Congress, pp. 27-36. FAO, Rome.
ITTO. 1996 Reduced impact, increased cost? Do reduced impact
logging regimes also reduce profits for forest operations?
Tropical Forest Update 6(3): 10-12.
SOUZA,
C.M. l999 A summary of research carried out under an
1TTO Fellowship to evaluate the use of GIS for the planning
of timber extraction in the eastern Amazon. Tropical Forest
Ul~date9(2): 24-25.
rain forest of Guyana: ecological, economic and silvicultural
consequences. Tropenbos-Guyana Series 6. The TropenbosGuyana Programme. Georgetown, Guyana.
24
International Forestry Review 2 ( l ) ,2000
Testing the applicability of reduced impact logging in
greenheart forest in Guyana
PETER VAN DER HOUT
l
'r2
Tropenbos-GuyanaProgramme, 12E Garnett Street, Campbellville, Georgetown, Guyana.
[email protected]
Utrecht University, Department of Plant Ecology, PO Box 800.48, 3508 TB Utrecht, The Netherlands.
SUMMARY
Selective logging in Guyana differs little from logging elsewhere in the tropics. Timber companiesharvest a small number of species without
much care for the remaining stand and hence future yields. Logging intensities tend to be low ( 5 m' ha-'), but in Chlorocardium rodiei
(greenheart) forests, high yields are obtained because harvestable stems tend to occur in clumps. Consequently, large gaps are formed in
the forest canopy. Skidder movements near the stumps destroy commercial regeneration and compact the soil, further affecting forest
recovery. A reduced impact logging (RIL) system was designed to redress these environmental problems. It was shown that conventional
greenheart logging (CL) is not likely to be sustainable. RIL was successful in reducing skidding damage by 65% (disturbed ground area)
and in reducing the average size of felling gaps by 40%. Total loss of canopy cover was not reduced due to group-wise felling in CL as
opposed to more scattered felling in RIL. The implementation of RIL did not lead to an increase in logging cost, despite CL having the
advantage of group-wise felling, because a higher yield was obtained per hectare and skidding was more efficient.
Keywords: cost-benefit analysis, greenheart, group-wise felling, Guyana, logging damage, reduced impact logging.
INTRODUCTION
Reduced impact logging (RIL) techniques aiming at
diminished logging damage have been recently introduced
in the tropics (Hendrison 1990, Bertault and Sist 1995,
Johns et al. 1996, Pinard and Putz 1996, Webb 1997). Such
logging techniques commonly consist of strict planning,
control of logging operations and training of a wellmotivated workforce (Heinrich 1995, Dykstra and Heinrich
1996). Planning includes pre-harvest stock inventory and
mapping, and topographic survey. Climber cutting, planning
and demarcation of skidding trails and directional felling are
also important components. High logging intensities can
seriously compromise the benefit of using RIL methods (Sist
et al. 1998). The country-wide introduction of RIL ultimately
depends on its financial merits. A financial comparison of
RIL with current practice is thus of great importance for the
acceptance of the former. Several studies have shown that
RIL is often just as cost-effective as conventional haphazard - logging (e.g.Mattson MBrn and Jonkers 198 1,
Hendrison 1990, Barreto et a1 1998, Holmes et al. 1999).
The main objective of the present study is to formulate
a logging concept under which timber can be extracted from
greenheart forest in Guyana, and similar forests elsewhere,
on a sustained yield basis (Van der Hout 1999). The specific
aims derived from this main objective are:
to describe the impact of the current logging practice on
the forest and its consequences for forest recovery and
future timber yields
to develop a reduced impact logging system that leaves
the forest in a condition that favours a rapid recovery to
a state that is silviculturally, ecologically and
economically desirable
to analyse the effect of logging intensity on forest
recovery and to determine at which intensity the benefit
of using reduced impact logging techniques, if any, starts
to be compromised
to examine the costs and benefits associated with a
change-over from customary practice to reduced impact
logging.
MATERIAL AND METHODS
Logging in Guyana
The tropical rain forest of Guyanais exceptional in the sense
that dominance by a few or one species occurs frequently
(Richards 1996). Chlorocardiurn rodiei (Lauraceae,
greenheart) occurs in such monodominant patches.
Consequently, relatively small proportions of timber
concessions are actually logged, yielding a low average
marketed volume per hectare, ca. < 5 m3 ha-'. However, in
areas of forest with high densities of commercial stems,
logging intensity can be as high at 20 trees per hectare
(Clarke 1956, Zagt 1997). Chlorocardiurn rodiei stands
occur mainly on gently undulating to rolling terrain with
Applicability of RIL in greenheart forest
slopes usually less than 20%. The soils are very deep and
well drained with topsoil textures varying between
unbleached sands, loamy sands, sandy loams and sandy clay
loams (Van Kekem et al. 1996).
Guyana has a legal diameter limit for cutting of 33 cm
dbh. Selective logging in Chlorocardium rodiei forests
typically takes all sound greenheart stems above 45 cm dbh
and, sometimes, in addition, smaller stems to be used as
pilings. Because of the gregarious spatial distribution of
greenheart, conventional practice (CL) leads to a distinctly
intermittent spatial distribution of large felling gaps
connected by skidding trails. Traces of skidder manoeuvring
in felling gaps because of positioning of logs for hooking are
common. The following environmental problems are linked
to CL (Brouwer 1996, Ter Steege et al. 1996):
i. the size of the felling gaps is ecologically and
silviculturally undesirable
ii. haphazard skidding seriously affects established
seedlings and saplings, and causes excessive soil
disturbance, both affecting forest recovery.
Development of a reduced impact logging system
As a baseline, it was decided to use a 'close to best practice'
operation instead of an unplanned, haphazard operation.
The latter type of operation is usually poorly organised and
equipped, and consequently inefficient, which hampers
insight into the effects of the technological innovations. The
CL operation featured a block layout (100 ha) and 100%
enumeration and mapping of harvestable trees (scale
1:2,500). Pre-harvest inventory was not used to plan skidding
trails or felling directions, but served mainly to quickly
relocate clusters of harvestable trees.
The RIL system features a string of sequentially related
activities, each depending upon the next one or the one
before. Besides block layout and 100% enumeration and
mapping of harvestable trees (scale 1: 1,250), liana cutting,
skidding trail planning, directional felling and winching
were applied. It also featured yield restrictions to discourage
formation of large felling gaps, that is, target trees were
distributed as evenly as possible over the area.
CL felling was carried out with a 13.0 kg Stihl 070,
whereas RIL felling used a 7.3 kg Stihl AV066. Aluminium
felling wedges (Ox-headTM) and a 0.8 kg sledgehammer
were used to facilitate directional felling. Both operations
used a Caterpillar 528 cable-arch skidder, in the case of RIL
equipped with separate choker straps with BardonTM choker
hooks. Both felling and skidding crews were expanded from
two to three persons.
The foremost reasons for felling a tree in a particular
direction are to ensure the safety of the felling crew and to
prevent damage to the bole of the felled tree but they do not
specifically apply to RIL alone. Using directional felling to
form multiple tree fall gaps was considered undesirable
because such felling gaps tend to be too large. Some authors
stress the potential of directional felling in avoiding damage
to potential crop trees (Pinard et al. 1995, Johns et al. 1996).
25
Others stress that the damage resulting from excessive
machine movements tends to be more serious and that
skidding damage can be more easily reduced than felling
damage (Mattsson-Mkn and Jonkers 1981, Whitman et al.
1997). It is clear that both cannot be optimised
simultaneously. In the present study, field observations
before logging had indicated that the density of potential
crop trees (dbh r 20 cm) was high, at some 5 0 trees ha-' ,
and that these trees were clustered around present crop trees
(Van der Hout 1999). Therefore, it was decided to aim at
reduction in skidding damage. For that purpose, trees were
felled at obtuse angles to the skid trails (1 35" to 150'). This
is the 'herring-bone system' (Conway 1982, Hendrison
1990). The felling direction was adjusted to guarantee the
safety of the felling crew and to prevent damage to the
harvested stem, if necessary.
It is generally suggested that with classic directional
felling techniques (cf Conway 1982, Klasson and Cedergren
1996) it is possible to fell a tree up to 90' from the direction
of a not overly heavy lean (Hendrison 1990). In order to be
able to fell against the natural lean, we used an innovative
technique described in Brunberg et al. (1984). This felling
technique leaves a small corner uncut, next to which a wedge
is inserted to stabilise the tree. This prevents the tree from
settling back on the guide bar when leaning backwards and
from falling before the felling cut is finished, whence
splitting, when leaning forwards.
Experimental lay-out
The study was conducted at Pibiri in Central Guyana. The
study considered CL with two logging intensities and RIL
with three logging intensities.
Conventional logging
Conventional logging was monitored in three 12 ha units
(200 X 600 m?) immediately bordering the RIL experiment.
In the centreof each 12 ha unit, a 20m wide, 500m long strip
was inventoried. Selection of trees to be felled was left to
the felling crew of the timber company. The logging
intensity thus varied along with the density of stems that
were regarded as worthwhile logging by the felling crew.
The spatial distribution of the stumps was used to delimit 2
ha plots (100 X 200 m') with logging intensities of 8 and 16
trees ha-' . No areas of this sizecould be found with alogging
intensity of 4 trees ha-'. In all, three replicates of both
intensities were demarcated.
Reduced impact logging
The RIL experiment also serves to gain insight into the effect
of logging intensity and post-harvest liberation on growth
and yield. A randomised block design is used for this
purpose. The experiment uses three blocks, each of which is
one replicate of the experiment. In each replicate, treatments
have been allocated to plots using a random selection
procedure. The experiment features four treatments: i.e. a
26
P. van der Hout
harvest of 4, 8 and 16 trees ha-l and a harvest of 8 trees
ha-' succeeded by a post-harvest liberation treatment
(poison-girdling). These intensities are based on average
maximum stock under present market conditions of 16 trees
ha-' , while 8 trees corresponds to ca. 25 m3ha -l, a volume
commonly found in literature on sustainable extraction rates
(e.g. De Graaf 1986).
The layout of and recording in the plots follows the
recommendations by Alder and Synnott (1992). Assessment
plots in the RIL area measure 1.96 ha (140 X 140 m2) with
a measurement plot surround of 50 m, implying a treatment
plot size of 5.76 ha (240 X 240 m2).
Damage assessments
In the RIL area, assessments were carried out before and
after skidding. Loss of canopy cover and tree damage due to
felling were recorded. After skidding, skidding trails were
mapped and additional tree damage was recorded. In the
conventionally logged area, assessments were only carried
out after skidding.
The size of felling gaps was estimated using Brokaw's
(1982) centre-point method, applied rather loosely (see
Brils and Laan 1995, c$ Van der Meer and Bongers 1996).
In assessing skidding damage, a distinction was made
between planned main trails, unplanned branch trails and
traces of manoeuvring in felling gaps and along trails. The
width of a skidding trail was the maximal extent of damaged
vegetation according to visual assessment, and was recorded
every tenth metre. The area occupied by trails was then
calculated by summing the areas of the resulting trapeziums.
Large disturbed patches, formed as a result of manoeuvring
in the felling gap or creating a 'ramp' for bundling, were
measured in the same manner as canopy openings, that is, by
the centre-point method.
Damage to individual trees r 10 cm dbh was recorded
in and around felling gaps, on and along the skidding trails,
and on and around manoeuvring areas. For each damaged
stem, we noted if it belonged to a commercial species, and
categorised the type of damage: cut (not for timber, but to
enable felling of a neighbouring timber tree), crushed,
uprooted, snapped, top broken off, split trunk, bark laceration
(width and length), crown damage (percentage excised), or
leaning or bending (deviation from vertical in both cases).
Tree survival was monitored for two years after logging.
Performance assessments
The performance of felling and skidding was determined in
m3per hour for both logging systems. The cost of felling and
skidding ($ m3)was determined by determining the hourly
cost ($ h-') of each activity and combining this with its output
(m3 h-'). The variation in stand characteristics, terrain, and
distances between trees implies that ceteris paribus did not
apply. Regression models were used to filter differences in
site quality, tree dimensions, extraction distances and skidder
load sizes. These models are based on detailed time studies
for felling and skidding and on standard rules of thumb for
the remaining phases of the logging process (Caterpillar
1986, FAO 1977, 1978), using data provided by the timber
company.
LOGGING DAMAGE
Overview of the logging operations
In the C L operation, the average dbh of the felled trees was
56.4 cm (sd=9.4 cm) and the average volume over bark was
3.2 m3 (sd=1.4 m3) per tree. The absolute logging intensity
varied over the area, in some parts amounting to 25 trees
ha-' , in other parts leaving the forest completely untouched.
The harvested gross volume thus varied between nil and 78
m3 ha-' . In this operation, 96% of the extracted trees were
Chlorocardium rodiei.
In the RIL operation, logging intensity was part of the
design. The average dbh of the felled trees was 55.7 cm
(sd=13.6 cm); the average log length 17.1 m (sd=3.7) and
the average volume over bark 3.4 m3 (sd=2.3 m' ) per tree.
One of the aims of the RIL concept was to reduce the size
of felling gaps. In order to achieve this, trees to be felled
were spread as evenly as possible over the area. This led to
a relief of pressure on Chlorocardium rodiei, its share being
reduced to 53% of the cut trees.
Loss of canopy cover
Accumulated canopy loss
Loss of canopy cover was correlated with logging intensity,
but differed along with the logging method. The relationship
between the felled basal area and the extent of accumulated
canopy loss reveals that reduced impact logging resulted in
a greater loss when the logging intensity was increased
above a level of 8 trees ha-' (see Figure 1). In the CL
operation, the average canopy opening per felled tree
decreased 23% when the intensity was increased'from 8 to
16 trees ha -'(Table 1). In the RIL operation, a much smaller
reduction by 4% was achieved with the same increase in
intensity. Fewer, but larger, gaps were formed in the CL
operation.
Differences between the logging methods in respect
of the size of the felled trees or the extent of natural gaps
prior to logging failed to explain this difference in trend
(Van der Hout 1999 - data not shown). According to the
tree selection method, felled trees were spaced out in
RIL, against clustered in CL. Because the local richness
of the forest determined the logging intensity with CL
and because a richer stand in practice meant larger
clumps of greenheart, gaps overlapped increasingly with
increasing logging intensity in CL. Gaps formed in RIL
overlapped less often than gaps formed with CL, and
overlapping increased only slightly with increasing
intensity.
Skidding effkiency
Regression analysis was used to build a model of the time
required for the extraction of a load under 'identical' (load
size, distance, etc.) circumstances. Owing to the small scale
of this study, it was not possible to evaluate loaded and light
travel times in detail. Therefore, the distance actually
travelled was estimated by empirical determination of the
'indirectness factor' of the two trail systems. A factor of 1.38
was derived for CL and of 1.14 for RIL. The average
distance travelled for a 1 km by 1 km drainage area - average
skidding distance as the crow flies 383 m - was thus
estimated at 527 m for CL and 436 m for RIL.
The model showed that implementation of the RIL
system reduced the time required to extract a load by 9%.
The travel time is shorter in RIL, but this is counteracted by
an increase in loading and unloading time of 16%. The latter
is not so much related to the technique as to the fact that trees
were felled in clusters in the CL operation.
Based on the time involved in a single round-trip, the
output of an eight hour working day (productive machine
time: 6 hours) was estimated. It appears that the
implementation of RIL increased skidding output from 14.4
m3 h -' to 15.9 m3 h - l .
LOGGING COSTS
The hourly output of felling was significantly reduced by
employing a directional felling method, which, however,
benefitted the performance of the skidding operation, for
which the hourly output was increased. Performance and
inputs expressed in time units need to be translated into
monetary units. Moreover, the comparison is not between a
method of felling and a method of skidding, but rather
between two logging systems.. Thus, for example, the
decision to fell a tree in a certain direction can only be made
after the inventory data have been processed and a trail plan
made. This implies that the cost-effectiveness of the entire
string of operations has to be considered instead of the costeffectiveness of each sub-operation in isolation.
Restricting the assessment of the operations until
unloading at the delivery point, the costing comprises:
29
Table 3 shows that use of the RIL system led to a
threefold increase in pre-harvest planning and preparation
costs and a twofold increase in felling cost, which were only
partly offset by lower skidding cost. The higher output per
hectare and per day reduced the cost per m3 of road
construction as well as indirect cost factors, such as costs of
logistics, support, area fees, etc. In the event these gains
neutralised the differences in the direct logging costs, and
the RIL system is thus neither necessarily more expensive,
nor cheaper than harvesting in the customary way.
TABLE3 Cost of logging of conventional and reduced
impact logging in Pibiri, Central Guyana
Cost in US$ m-'
Operation
Pre-harvest planning
Conventional
Reduced impact
0.18
0.50
Harvest preparation
0.52
0.56
Felling and cross-cutting
0.60
1.16
Skidding
4.30
4.10
Landing operations
0.34
0.32
12.18
12.18
Road maintenance
1.91
1.64
Support, logistics and
supervision
3.37
2.90
Trucking, loading and
unloading
Other overhead costs
1.99
1.99
Royalty and area fee
2.89
2.88
28.29
28.23
Total
Notes: costs based on logging intensity of 10 trees ha-'
(conventional yield 28.5 m3 ha-', reduced impact yield=31.0 m?
ha1),feeder road density of 1 km km-*;average skidding distance
as the crow flies of 383 m and hauling distance 50 km. Daily output
based on skidding performance; i.e. 84 m3d.' for conventional and
97 m3 for reduced impact logging.
DISCUSSION
Logging damage
pre-harvest planning
harvest preparation
stump operation
off-road transport
landing operations
road transport, loading and unloading
to which are added the costs of:
transport routes
overheads.
Unfortunately, it was not possible in the present study to
carry out a detailed study on the performance and cost of
road transport. Therefore, haulage costs were assessed on
the base of random field observations and rules of thumb
(FAO 1977, 1978, Caterpillar 1986).
Comparison of results of logging damage studies
Implementation of RIL techniques has been reported to
result in spectacular reductions in damage to the remaining
trees. Bertault and Sist (1995), Hendrison (1990), Johns et
al. (1996) and Pinard and Putz (1996) reported a reduction
of residual stand damage by around 20% or more. Mqjor
reductions in canopy removal and skidding trail coverage
have been reported as well. In the present study, RIL resulted
in a much more modest reduction in residual stand damage
(Figure 4). The achievements are still poorer when comparing canopy loss (Figure 5). In fact, only the reduction of the
skidding trail coverage is in agreement with the results of
other comparative studies.
30
P. van dei. Hout
Careful examination of Figure 4 reveals that intensityadjusted residual damage was lower in the CL operation in
Guyana than in most RIL operations elsewhere. The
spectacular damage reductions in other studies should be
seen against a higher benchmark. This could be interpreted
as indicating that relatively little damage is inflicted by CL
in Guyana. However, damage levels in Guyana are consistent
with findings in two studies in the Brazilian Amazon (Uhl
and Vieira 1989, Verissimo et al. 1995), Costa Rica (Webb
1997), French Guiana (Schmitt & Bariteau 1990) and
Suriname (Jonkers 1987), indicating that such an interpretation would not be justifiable. Apparently differences
occur between continents; smaller tree sizes in the neotropics
may play an important role.
FIGURE
4
Comparison between logging operations of total
residual stem damage estimates for trees with dbh 2 1 0 cm.
Damage is defined as the sum of destroyed and injured residual
stems. The relationship between the percentage of residual
stem damage and logging intensity is highly significant, but
explains only a moderate amount of the total variation: (%
canopy loss)=13.16+1.96 (number of trees removed ha-');
R2=0.49.
Studies used in this comparison were conducted in Para
State, Brazil (Bzl: Uhl and Vieira 1989; Bz2:Verissimo et al.
1992; Bz3: Verissimo et al. 1995; Bz4: Johns et al. 1996),
Costa Rica (CR: Webb1997),Gabon (Gb: White1994), French
Guiana (FG: Schmitt and Bariteau 1990), East Kalimantan,
Indonesia (EKI: Abdulhadi et al. 1981; EK2:Bertault and
Sist1995), West Malaysia (WM:Johnsl988),Sabah, Malaysia
(Sbl; Nicholsonl979; Sb2: Pinard and Putz1996), Sarawak,
Malaysia (SW: Mattson MHrn and Jonkersl981), Suriname
(Su:Jonkersl987), and the present study (Gy)
Recommendations for sustainable logging
Previous research has indicated that the maximum allowable
canopy opening to avoid adverse ecological and silvicultural
effects is 500 m' (Brouwer 1996), while Ek and Van der
Hout (in prep.) found changes in species composition in
gaps over 300 m2.Gap (cluster) sizes found with CL were,
on average, larger than the latter figure, even at the lower
logging intensity. Implementation of RIL led to gap sizes,
on average, below 300 m2, as long as the intensity did not
exceed 8 trees ha-'.
Gap size increased proportionately less when the logging
intensity was increased in case of CL. This is plausibly
explained by a difference in felling pattern and spacing
between felled trees. An increase in logging intensity meant
an increase of the number of trees in a cluster in case of CL,
i.e. an increased overlap in felling gaps. In case of RIL, this
meant only a marginal increase in overlap of felling gaps, but
an increased number of interlinked gaps.
Proper planning of harvest activities, pre-alignment of
skidding trails, a herring-bone felling pattern and winching
reduced the area traversed by the skidder considerably.
The reduction could mainly be ascribed to an almost
complete elimination of traces of skidder movements in
felling gaps.
The following conclusions can be drawn from the
results:
FIGURE
5
Comparison of canopy loss estimates as a
function of logging intensity ( d h a ) for conventional and
reduced impact logging operations in several sites. The
relationship between the loss of canop)] cover and logging
intensity is signrficant, but explains a meagre proportion of the
total variation; (% canopy loss) = 12.79+ l . 13 (number of trees
renfoved ha-'); R2=0.22, F1,17=4.90, pc0.05.
Studies used in this comparison were conducted in North
Queensland, Australia (Au: Crome et al. 1992), Park State,
Brazil (Bz1:Uhl & Vieira1989; 822: Verissimo et al. 1992;
Bz3: Verissimo et al. 1995; Bz4:Johns et al. 1996), Costa Rica
(CR: Webb1997), Gabon (Gb: White1994), French Guiana
(FG: Schmittl989), Suriname (Su: Jonkersl987). Sarawak
(SW: MattsonMdrn and Jonkersl981), and this study (Gy)
conventional logging of greenheart-bearing forest stands
leads to the formation of large canopy openings that
involve high and undesirable silvicultural risks
lowering the logging intensity below 8 trees per hectare
in RIL does not reduce the average felling gap size, while
raising the intensity to 16 trees per hectare leads to
silviculturally undesirable gap sizes
in the present study, no difference was found in the size
of single tree fall gaps. This suggests that liana cutting
and directional felling as carried out in this study did not
reduce the amount of canopy lost
implementation of RIL elements reduced the area
traversed by the skidder by about two-thirds depending
on the logging intensity, while skidder movements in
felling gaps were reduced by about three-quarters
depending on the logging intensity.
Logging efficiency
Several studies have shown that RIL does not have to affect
the cost-effectiveness of a logging operation, because
additional costs incurred during planning and directional
felling were more than compensated by increased efficiency
of the skidding operation. Mattson-Mirn and Jonkers (198 1)
reported additional costs due to planning and pre-opening of
skidding trails amounting to 4% of the total cost of
traditional skidding, whereas actual skidding cost was
decreased by 27% in Sarawak. Hendrison (1990) reported a
reduction of direct costs by 16%. Barreto et al. (1998)
reported that felling in a 'planned' operation was 18% less
productive than felling in an 'unplanned' operation, whereas
skidding performance was increased by about 27% in the
Paragominas region in Brazil. Also in Brazil, Holmes et al.
(1999) found that pre-harvest planning and preparation
costs in a RIL operation were eleven times the costs incurred
during this phase in a CL operation, and that the cost of
felling was 26% higher. There was a gain in skidding
efficiency which amounted to acost reduction of 42%. Costs
of road construction, landing operations and of overheads
were also reduced by implementing RIL, resulting in an
overall cost reduction of 5%.
In a study by Winkler (1997) in the eastern Amazon, preharvest planning and preparation did not occur in the
traditional operation; felling cost was 3 1% higher in the RIL
operation, whereas skidding cost was 33% lower. In this
study in the eastern Amazon, the logging intensity in the RIL
operation was substantially lower than in the CL. This
resulted in the aggregate cost per m'at the landing being 9%
higher in the RIL system than in the traditional operation.
Why do the results of the present study differ from those
of the other studies? Skidding efficiency was enhanced by on
average 33% in most other studies, and felling efficiency
declined by on average 30%. In the present study, skidding
efficiency was only marginally improved, whereas felling
efficiency declined substantially. There are several reasons:
differences of a technical and organisational nature
existed between the logging operations compared
elsewhere
in the present study, the CL operation featured felling in
groups, whereas trees were scattered in the RIL
operation; this feature gave the CL operation an
advantage
in the present study, a special felling technique was
employed in the RIL operation, which to the best of our
knowledge was more sophisticated than the directional
felling techniques used in the other RIL operations.
The tree selection criteria adopted clearly reduced the
financial attraction of RIL. Partly this was related to the
higher efficiency of group-wise felling, but another, more
important, factor is the change in species composition of the
output. Undoubtedly, the aspect of foregone timber reduction of the proportion of Chlorocarclium rodiei from
96% to 53% of the coupe - will be the most difficult one to
reconcile with the interests of the logger. In the end, it
31
depends on the forest policy of Guyana and on the
expectation that conventional tree selection methods will
not qualify for certification. Implementing the tree selection
criteria will bring good forest management closer and most
likely will ensure access to privileged markets, perhaps with
bonus prices.
ACKNOWLEDGEMENTS
The investigations reported in this paper were carried out at
the Pibiri fieldwork site of the Tropenbos-Guyana
Programme. This study was made possible through funding
provided by the Tropenbos Foundation, Wageningen, the
Netherlands and Demerara Timbers Limited, Mabura Hill,
Upper Demerara, Guyana. Many people contributed
significantly to the collection of data used in this paper. In
this respect, my thanks go foremost to the Pibiri field crew
and to the Dutch and Guyanese students who participated in
the project - especially Chris Brils and Edwin Laan who
carried out most of the RIL assessments.
REFERENCES
ABDULHADI,
R., KARTAWINATA,
K.and. SUKARDJO,
S. 1981 Effects
of mechanized logging in the lowland dipterocarp forest
at Lempake, East Kalimantan. Malay. For: 44 (28~3):
407-416.
D. and SYNNOTT,
T.J. 1992 Permanent sample plot
ALDER,
techniquesfor mixed tropical forest. Tropical Forestry Paper
25. Oxford Forestry Institute, Oxford, U.K.
BARRETO,
P,, AMARAL,
P,, VIDAL,
E. and UHL.C. l998 Costs and
Amazonia. For: Ecol. Manage. 108: 9-26.
BERTAULT.
J.-G and Srsr.,P. 1995 Impact de I'exploitation en for&
naturelle. Revue Bois et Forgts des Tropiques 245: 5-13.
BRILS,
C A J , and LAAN,
E.A. 1995. Gaps and damage inflicted by
directional felling in a tropical rain forest in Guyana M.Sc.
thesis, Department of Forestry, Wageningen Agricultural
University, Wageningen, The Netherlands.
BROKAW,
N.V.L. 1982 The definition of tree fall gap and its effect
on measures of forest dynamics. Biotropica 14: 158-160.
BROUWER,
L.C. 1996 Nutrient cycling in pristine and logged
tropical rain forest; a study in Guyana. PhD thesis, Utrecht
University. Tropenbos-GuyanaSeries 1,Georgetown, Guyana.
BRUNBERG,
B., GARDH,
R. and LINDGREN,
P. 1984 Felling manual.
Forskningsstiftelsen 'Skogsarbeten' (Forest Operations
Institute), Stockholm, Sweden.
CATERPILLAR 1986 Caterpillar performance handbook. 17th
edition. Caterpillar Tractor Co., Peoria, IL, USA.
CLARKE,
E.C. 1956 The regeneration of worked-out greenheart
(Ocotea rodiaei) forest in British Guiana. Emp. For: Rev. 35:
173-183.
CONWAY,
S. 1982 Logging practices: principles of timber
harvesting systems. 2nd edition. Miller Freeman Publications,
San Francisco, USA.
L.A. and RICHARDS,
CROME,
F.H.J., MOORE
logging damage in upland rain forest in north Queensland. For:
Ecol. Manage. 49: 1-29.
DE GRAAF,N.R. 1986. A silvicultural system for natural
regeneration of tropical rain forest in Suriname. PhD thesis,
International Forestry Review 2 ( l ) ,2000
33
Lessons learned from the implementation of reduced-impact
logging in hilly terrain in Sabah, Malaysia
MlCHELLE A PINARD1, FRANCIS E PUTZ2and JOHN TAY3
l
University of Aberdeen, Department of Forestry, University of Aberdeen, 581 King Street, Aberdeen AB24
5UA, U.K. [email protected]
Center for International Forestry Research (CIFOR), PO Box 6596, JKPWB, Jakarta 10065, Indonesia and
Department of Botany, University of Florida, P 0 Box 118526, Gainesville, FL 3261 1-8526, U.S.A.
Rakyat Berjaya Sdn Bhd, lnnoprise Corporation, P 0 Box 11622, 8881 7 Kota Kinabalu, Sabah, Malaysia.
SUMMARY
Between 1992 and 1997, about 2400 ha of old growth dipterocarp forest in southeastern Sabah was logged according to reduced-impact
logging (RIL) guidelines as part of a pilot carbon offset project. Harvest planning, vine cutting,directional felling, and skidding restrictions
contributed to a reduction in stand damage from 50% to 28% of the original stems; damage to soil was reduced from 13% to 9% of total
area in RIL relative to conventional logging areas. Residual stands in RIL areas had greater vertical structure and better stocking of
commercial timber species than stands in conventionally logged areas, with positive gains for conservation of biodiversity and sustainability
of timber production. Steep terrain and the lack of predictable dry periods were constraints on the ground-based skidding system, and
resulted in large volumes of timber being inaccessible, and in production delays caused by wet weather. Introduction of an aerial yarding
system in this region would allow a greater proportion of the areas to be harvested in an environmentally acceptable way.
Keywords: logging damage, soil disturbance, steep slopes, stocking, wet weather shutdowns.
INTRODUCTION
Concern over unnecessarily destructive timber harvesting
practices and rising levels of anthropogenic greenhouse gas
emissions stimulated the development of a reduced-impact
logging (RIL) project in Sabah in 1992. The aim of the
project was to reduce damage to soils and the residual forest
by one half, relative to local conventional harvesting
operations. By reducing incidental tree mortality, less
biomass would be lost from the forest, carbon emissions
from decaying debris would be reduced, and the capacity of
the forest to sequester carbon would be maintained. The
initial project (1400 ha) was funded by a US-based power
company as apilot carbon offset project, where the potential
implications of damage reductions for greenhouse gas
emissions could be explored (for discussion of the rationale
for the carbon offset see Putz and Pinard 1993, Pinard and
Putz 1996, Pinard and Cropper (in press). The project was
extended to an additional 980 ha in 1996, funded by a
consortium of power companies also interested in the offset
potential of RIL.
The key environmental problem that the project was
intended to address was excessive damage to soils and
advanced regeneration during conventional harvesting
operations. On average when old growth dipterocarp forests
in Sabah are logged, 8-15 trees are extracted per ha,
representing 50-1 50 m3of timber. As many as 40-70% of the
residual trees are damaged (Fox 1968, Sabah Forestry
Department 1989) and 15-40% of the area traversed by
bulldozers (Chai 1975, Jusoff 199 1). Little pre-harvest
planning is done, available topographic maps are unreliable,
and lack of co-ordination between chainsaw and bulldozer
operators results in inefficient and damaging operations
(Pinard et al. 1995). Current practices in Sabah are not
sustainable because the volumes of timber extracted, the
area logged each year, and damage to advanced regeneration
are all too high (Sabah Forestry Department 1989). Although
a new silvicultural system is needed (Udarbe et al. 1994),
this project's focus was on harvesting only.
PROJECT AREAS
The project is based in old growth dipterocarp forest in three
forest reserves in southeastern Sabahi. The forests are
diverse in tree species (Fox 1978, Newbery et al. 1992), and
heavily stocked with commercial trees; average basal area
ranges between 25 to 33 m? ha-I (trees 210 cm dbh), about
68% of this is in commercial species (Pinard and Putz 1996).
Mean canopy height is 45 m and emergent trees reach 70 m.
The terrain consists of a series of short (200-300 m long),
steep ridges. On average, only 22% of the project areas had
'
The project areas include three parcels: 450 ha in Ulu Segama,
5" 0' N, 117" 30' E, 150 - 750m asl), 950 ha in Kalabakan (4"
25' N, 117'29' E, 150 - 900 m asl), 980 ha in Gunung kara (5'
0' , 117" 30' E, 150 -750 m asl).
34
M.A.Pinard, F.E. Putz and J. Tay
slopes less than 20 degrees; 46% had slopes 20-34 degrees
and 29% had slopes greater than or equal to 35 degrees
(Table 1). Underlying geological substrates and soils are
variable (Nussbaum 1995). Rainfall is aseasonal, with mean
annual rainfall at the three forest reserves ranging from
2700-3100 mm yr' (estimated from Walsh 1996).
TABLE1 Percent of study areas in slope classes. The two
logging treatments were reduced-impact logging (RIL) and
conventional logging (CL)
Reserve
Ulu Segama
Kalabakan
Slope
RIL
CL
RIL
<loo
10-19"
20-29"
30-34"
35-39"
-> 40"
9%
20%
28%
16%
1%
13%
38%
22%
22%
6%
2%
15%
32%
20%
14%
16%
10%
16%
CL
21%
30%
34%
10%
2%
3%
Gunung Rara
RIL
5%
15%
26%
22%
15%
17%
Notes: estimates are based on slope measurements taken in
permanent plots (1600-3200 m?) distributed in a stratified
random manner in study areas. Sample sizes were as
follows: Ulu Segama, RIL N=138, CL N=120; Kalabakan,
RIL N=127, CL N=125; Gunung Rara, RIL N=86.
The project is managed by Rakyat Berjaya, the forest
products subsidiary of Innoprise Corporation Sdn Bhd
(ICSB), the commercial arm of the Sabah Foundation
(Yayasan Sabah). The project sites are within the Sabah
Foundation timber concession. The state forestry department
issues annual cutting licences and harvesting operations are
largely subcontracted out to private companies. All pre- and
post-harvesting silvicultural operations are conducted by
Rakyat Berjaya. The reserves do not have human
populations living within their borders
The conventional timber harvesting system used in
Sabah since the mid- 1960s is based on a minimum harvesting
diameter of 60 cm dbh, and bulldozers are used to make
skidding trails and to extract logs (Sabah Forestry
Department 1989). The management system is a modified
uniform system with 60 year cutting cycles, but is currently
under review and revision (Kleine and Heuveldop 1993).
'Qpically, the only silvicultural activities implemented,
aside from harvesting, are a pre-harvest inventory (2.5%)
and post-harvest regeneration sampling. Although other
silvi-cultural treatments (e.g. pre- and post-harvest climber
cutting, girdling of non-commercial stems) were included as
prescriptions in the original system, these treatments were
dropped in the mid-1970s because high levels of damage
associated with harvest meant that only a small proportion
of the residual forest warranted treatment (Chai and Udarbe
1977). Generally at the time of harvest, some advanced
regeneration is present. Dipterocarp seedling densities tend
to be variable both spatially and temporally.
Rn. areas were harvested by crews trained in reduced
impact logging techniques. The Ulu Segama parcel was
harvested in 1993- 1994, the Kalabakan parcel was harvested
in 1995-1996, and the Gunung Rara parcel was harvested in
1997-1998.
The project maintains a rigorous monitoring programme
for measuring impacts on residual trees and soils. The
monitoring programme used in the first project site (Ulu
Segama) was designed to address several objectives: l ) to
provide reliable estimates of damage to soils and residual
trees in RIL and conventional logging (CL) areas as
measures of project effectiveness; 2) to provide data to
document changes in standing stocks of carbon in RIL and
CL areas over time; and, 3) to provide growth and yield data.
The programmes used in the second (Kalabakan) and third
(Gunung Rara) sites were designed to address the first
objective only. The rangers also keep detailed records of all
operations to facilitate re-entry for the next cut and to
provide information on contractor compliance with the
guidelines.
DESIGN OF MONITORING PROGRAMME
The basic design of the monitoring programme involves preand post-harvest measurements of stand conditions in project
areas (i.e.to be logged according to RIL guidelines) and in
adjacent areas to be logged using conventional practices.
Prior to logging in each of the first two project sites,
permanent plots were established at a 10% intensity in four
logging units randomly selected from RIL and four from CL
areas. Units in the RIL and CL areas were paired according
to topography and logging schedule to reduce variability in
logging impacts due to soil moisture and slope. Plots were
located in a stratified random manner. In the third project
site, sampling intensity was dropped to 2.5% (3.2% in RIL,
1.6%in CL) in order to reduce costs. At this site, no logging
unit divisions were used and plots were located in a stratified
random way throughout the 980 ha project area and the
adjacent 1000 ha CL area2.
Trees in the plots were tagged, mapped, and diameters
measured at breast height (at 1.3 m or above buttresses,
hereafter, dbh). For commercial trees, species or species
group was recorded. Any existing damage to trees was
recorded. Plots were reassessed for damage 5-30 days postharvest. Damage was recorded by type (e.g. stem, crown,
bark, root) and intensity (5 point scale). In the first project
site, plots were again resurveyed 10-12 months post-harvest
to measure tree mortality and again three years post-harvest
to measure growth, recruitment, and mortality. Summary
pre- and post-harvest data are presented in Table 2.
In each of the first and second project sites, soil
disturbance was mapped and measured at 100% intensity in
the eight logging units (four RIL, four CL). In the third
project site, soil disturbance was estimated from sampling
along transects located in a stratified random way. A
summary of results is presented in Table 3.
the time of publication, the data from the CL area are
unavailable.
At
Implementation of RIL in hilly terrain, Sabah
35
TABLE2 Pre-logging conditions and timber extraction data from 3 reduced impact logging (RIL)and 2 conventional logging
(CL) area in three forest reserves
Reserve
Ulu Segama
Kalabakan
RIL
CL
Gunung Rara
RIL
RIL
CL
Pre-logging (no. ha-')
harvestable trees
23.3 (4.4)
21.7 (2.3)
19.5 (4.5)
18.0 (2.1)
16.6 (7.8)
Pre-logging basal area
trees 210 cm dbh (m2ha-')
27.4 (3.4)
27.5 (4.3)
26.4 (4.0)
23.2 (3.8)
26.5 (0.7)
8.8 (3.6)
13.6 (2.7)
8.4 (2.1)
11.2 (1.1)
5.3 (8.6)
103 (54)
152 (23)
81 (20)
111 (7)
50 (49)
60 (30)
152 (23)
48 (20)
111 (7)
27 (9)
Area unlogged
(% of total)
Trees (no. ha-')
extracted from net area logged
Volume (m3ha-')
extracted from net area logged
Volume (m3ha")
extracted from entire block
Notes: means presented with standard deviations noted parenthetically. Estimates from Ulu Segama and Kalabakan are based on mean
values of four logging units (50-70 ha), subsampled with 15-35 plots per unit. The estimates from Gunung Rara are based on 87 plots
distributed in a stratified random way throughout the entire 980 ha. Harvestable trees are trees of commercial species, >60 cm dbh.
Estimates expressed per net area logged are based on a sub-sample of the plots, excluding plots located in areas that were not logged. More
detail on methodology can be found in Pinard and Putz (1996).
TABLE3 Soil disturbance as a proportion of total area in the study areas. The two logging treatments were reduced-impact
logging (RIL) and conventional logging (CL)
Reserve
Ulu Segama
RIL
Kalabakan
CL
RIL
CL
Gunung Rara
RIL
Area with disturbed soil
6.8% (2.6)
16.6% (2.3)
9.5% (5.1)
10.2% (1.5)
9.3% (8.3)
Roads and landings
3.38 (2.5)
4.7% (0.8)
0.4% (0.6)
Not available
1.2% (1.6)
Skidding trails (total)
3.5% (2.1)
11.9% (2.7)
9.1% (4.7)
8.0% (1.1)
6.9% (6.2)
Notes: values presented are mean percentage of total area with standard deviation noted parenthetically. Sample size in Gunung Rara was
21 transects of variable length, totalling 3.2 km. Sample size in RIL Kalabakan was 8-10 transects of lOOm in each of three logging units.
DEVELOPING RIL GUIDELINES
The motivation for the project initially came from the Sabah
Foundation's interest in carbon offsets for gaining funds for
improving forest management. As the pilot project was
being developed, an initial set of guidelines was drafted
based on best management practices recommended in
Indonesia, Malaysia and Australia. The guidelines have
been reviewed biannually and revised to make them more
appropriate to local conditions. The rangers working in the
field have been instrumental in the improvement of the
guidelines. Throughout the project, they have worked to
increase the efficiency and effectiveness of harvesting and
developed a record-keeping system to document progress.
(For more details on implementation of the project and
development of the harvesting guidelines see Pinard et al.
1995). The logging crews and forest rangers working in the
first experimental area were trained by foresters from the
Queensland Forest Service and by expert fellers from Sweden.
Subsequent crews and rangers involved in the project have
been trained by personnel within the project and ICSB.
To increase the project's transparency and credibility, an
independent committee was set up to assess compliance with
the guidelines. The Environmental Audit Committee has
three members, one appointed by the timber concessionaire, one by the power company and one joint appointment.
The committee meets about every 6 months to review
progress and to conduct an audit. Any changes in the
harvesting guidelines must be approved by the committee.
The committee reviews the results from the monitoring
programme and the rangers' records but conducts its own
field assessment to determine compliance. The involvement
36
M A . Pinard, F.E. Putz and J. Tay
of the committee provides a mechanism for discussing and
reviewing issues related to implementation of the guidelines.
Local research efforts have been directed at measuring
environmental and ecological impacts (e.g.Pinard and Putz
1996, Howlett 1998, Pinard et al. in press), costs and
benefits (Tay 1999), and implications for carbon storage
(Pinard and Cropper 2000), hydrology (R. Walsh and W.
Sinun unpubl.), silviculture (M. Pinard unpubl.) and biodiversity (Davis in review).
Closing operations include the removal of temporary
stream crossings and installation of drainage structures on
roads and skidding trails. The guidelines call for functional
draining cross drains on skidding trails on slopes, with a
recommended density of cross drains related to slope and
fetch.
LESSONS LEARNED
Effectiveness of RIL regimes in reducing damage
KEY FEATURES OF RIL PROCEDURES
The guidelines include specifications for pre-harvest
planning, vine cutting, felling, skidding and post-harvest site
closure. A 100% stock map (1:5,000 scale) is prepared
which shows all harvestable trees, stream and road buffer
zones, and sensitive areas that are to be excluded from
logging. Furthermore, during stock mapping rangers note
topographic features that are not apparent on the 1:50,000
maps used for harvest planning. Map improvements have
often proved useful when laying out roads and major skid
trails, but the cost savings in construction and maintenance
are difficult to estimate. Stock map preparation is costly
(about 40% of additional direct operational costs, Tay 1999)
but the rangers involved in the project have repeatedly
argued for their utility, particularly in increasing their
familiarity with the forest and for their application for future
harvests.
Vine cutting is conducted at least one year prior to
harvest and vines 22 cm dbh are cut in areas to be harvested.
Efforts are made to avoid cutting Ficus, other root climbers,
and other species of known importance to wildlife but this
requires training in identification that has sometimes not
been forthcoming. If, prior to vine cutting, efforts were
made to identify areas deemed unloggable due to slope,
proximity to streams, or other reasons, potential negative
impacts of vine cutting could be reduced, as could costs.
However, to date rangers have been reluctant to identify
these areas prior to the initiation of vine cutting.
Bulldozers are required to stay on marked primary and
secondary skidding trails and operators are encouraged to
use their winches to bring the logs to the bulldozer. Use of
the bulldozer blade on skidding trails is prohibited on slopes
c20 degrees; on slopes 220 degrees, the blade may be used
in skidding trail construction only. Skidding trail width is
restricted to a 4.5 m working surface (5 m if slope 220
degrees). Skidding is prohibited on slopes 235 degrees and
although no explicit wet weather restriction is stated in the
guidelines, the philosophy of minimising damage restricts
the use of heavy machinery when soils are saturated.
Skidding trails are marked on the harvest plan and in the
field, with end points clearly indicated. Trees to be felled are
marked with an " X and a 1 m long painted line to mark the
direction of intended fall. Potential crop trees (trees 215 cm
dbh of commercial species) that occur near trees to be felled
or skidding trails are marked with a blue ring to increase
their visibility.
At all three project sites, implementation of the guidelines
resulted in a substantial reduction in damage relative to
conventional practices (Figure 1). Damage to soils was also
less in RIL areas relative to CL areas, both in terms of area
damaged (Table 3) and degree of disturbance to topsoils
(Pinard et al. in press). Volumes extracted from the RIL
logging units have been, on average, lower than in CL areas3
(Table 2). The reductions in logging damage recorded in
RIL areas, however, cannot be attributed to lower harvesting
intensities alone. When the variability associated with
extraction levels is removed, stand damage in CL areas
substantially exceeds that in RIL areas. Also, the slopes of
the lines representing the relationship between basal area
extracted and proportion of basal area killed are similar for
the two logging methods; the intercept, however, is greater
for CL than RIL areas (Figure 2), meaning that for a given
0 ,
,
uprooted and crushed
m Snapped crown
,
,
,
0,
e r
e o rnI
FIGURE1 Logging damage in reduced-impact logging (RIL)
and conventional logging (CL) areas in Sabah. Bars
represent mean values from 4 logging units (50-70 ha),
subsampled with 15-35plots; plots in unlogged areas were
not included in the analysis
' Nevertheless, mean volumes extracted from RIL areas, when
expressed per net loggable area, are within the range of the
concession average of 70 m3ha?
38
M.A.Pinard, F.E. Putz and J. Tay
vegetation types are of particular concern because of the
potential for delaying forest recovery. Small modifications
to the tree selection rules could reduce this problem. For
example, the number of trees to be felled in groups could be
restricted, and the likelihood of vine or bamboo invasion in
an area could be considered.
Slope considerations for harvesting
The areas of old growth forest that remain within the Sabah
Foundation's concession tend to have rugged topography
andunstable soils (Table 1). These slopes have been a major
constraint for ground-based log extraction. Restrictions
against the use of bulldozers on slopes >35 degrees is
justifiable on environmental grounds, indeed many best
management practices prohibit heavy machinery on slopes
>20 degrees. However, because operators working in CL
areas were not working with a slope restriction, the restriction
on logging on steep slopes was associated with substantial
volumes of foregone timber. In the three project areas, about
43-47% was unlogged due to steep slopes or unstable soils.
Although the timber can be harvested when the appropriate
technology becomes available, this foregone timber
represents a high opportunity cost for the concessionaire
and logging contractor.
Converting to an aerial based yarding system in steep
and rocky areas seems the obvious solution to the problem
of inaccessible timber. Apparently viable options include
helicopter yarding and skyline operation. Helicopter yarding
would require little training of local people as it would be
necessary to employ a helicopter yarding company.
Currently, the yarding costs are thought to be prohibitive but
the use of helicopters in Sarawak suggests that it might be
worth investigating costs further. Skyline systems are less
costly per cubic metre yarded to the roadside than helicopter
yarding and given the Forestry Department's current positive
experiences with skyline yarding in Deramakot Forest
Reserve, this seems a more feasible option. A combination
of bulldozer and skyline yarding may be the most costeffective. Investment in a new yarding system would be
more attractive if there were some assurance that the newly
trained contractor (and operators) would remain working
with the concessionaire.
Within the project some tests have been made of the
effect of relaxing the guidelines to allow the bulldozers
access to steep slopes but otherwise follow the existing RIL
guidelines (Kalabakan). Trials conducted to date have
generally resulted in an unacceptable amount of soil
disturbance in areas where skidders operated and on roads,
both in terms of degree and extent. Furthermore, the steep
skidding trails were difficult to drain and therefore became
gullied and unlikely to be reusable.
Wet weather constraints
The lack of seasonality in rainfall meant that work in RIL
areas was often delayed; on average there were 18-25 rain
days per month. Again, trials with wet weather skidding
resulted in deep 'box cuts', trails that could not be drained,
increased soil compaction and subsoil disturbance. Use of
an aerial yarding system would reduce this problem.
Construction of proper drainage structures on skidding
trails was facilitated by requiring use of a small backhoe (or
excavator). Functioning cross drains were difficult to
construct using bulldozers, partly because of the need to
manoeuvre the bulldozer without causing additional damage.
In many cases, it was necessary to employ a worker to
manually open part of the cross drain to ensure that it would
drain. In areas of skidding trails and roads where deep box
cuts had formed (mostly due to skidding during wet weather),
it was very difficult to install any drainage structures. Water
bars or 'bumps' were used to slow water flow but in most
cases, the bars were breached within a couple of weeks
resulting in severe erosion.
Extending cable winching capacity could also help to
reduce problems with soil disturbance and slope restrictions.
In general the bulldozers operating in the RIL areas have
32mm winch cables that are about 30 m long. The cable is
heavy and difficult to drag distances of more than 10-12 m.
Recommended options for increasing winching distance
included a change to a lighter cable or the use of a small
winch to pull out the main cable. The problem was
ultimately solved by employing a second worker to assist in
setting the cable. Other improvements on skidding that have
been suggested include the use of chokers, an arch, or
skidding pans but, to date, none have been adopted by the
contractor.
CONCLUSION
Implementation of the RIL guidelines resulted in reductions
in stand damage from 50% to 28% of the original stems, and
soil damage from 13% to 9% of the total area in RIL relative
to CL areas. Timber volumes extracted in RIL areas, when
expressed per net area logged, were about 70% of the
volumes extracted in CL areas. While buffer zones near
streams and sensitive areas are responsible for part of the
reduction, the larger part of the reduction is probably due to
a tendency of fellers working in RIL areas not to fell trees
likely to cause extensive damage.
Post-harvesting stand conditions suggest that the benefits
for silviculture and biodiversity conservation are potentially
great. Higher densities of large trees in RIL areas provide
the vertical structure important for certain species. The
higher stocking levels of trees 10-40 cm dbh in RIL areas are
expected to result in higher commercial volume increments
over the next cutting cycle.
Wet weather shutdowns delayed operations and
restrictions against skidding on steep slopes made a
proportion of each project area 'unloggable' using the
current ground-based system. Introduction of an aerial
system, or a hybrid system where bulldozers are used in
combination with a skyline system, could help to reduce the
opportunity costs associated with minimising logging
damage in this region.
Implementation of RIL in hilly terrain, Sabah
ACKNOWLEDGEMENTS
We acknowledge the contribution of the silviculture unit and
the forest rangers of Rakyat Berjaya Sdn Bhd for their role
in monitoring the impacts of the project, and in supervising
implementation of the guidelines in the field. Pinard thanks
the Economic Planning Unit of the Government of Malaysia
for allowing her to conduct research in Malaysia.
REFERENCES
CHAI,D. N. P. 1975 Enrichment planting in Sabah. Malay. For
38: 271-277.
CHAI,D. N. P. and UDARBE,
M. P. 1977 The effectiveness of
current silvicultural practice in Sabah. Malay.For 40: 27-35.
DAVIS,
A. J. In review. Does reduced-impact logging help preserve
biodiversity in tropical rainforests? A case study from Borneo
using dung beetles as indicators. Ecological Entomology.
Fox, J. E. D. 1978 The natural vegetation of Sabah, Malaysia. l .
The physical environment and classification. Trop. Ecol. 19:
218-239.
HOWLETT,
B. 1998 Pioneer trees and forest recovery after logging
in Sabah, Malaysia. PhD Dissertation, University of Utah, Salt
Lake City, Utah, USA.
JUSOFF,K. 1991 A survey of soil disturbance from tractor logging
in a hill forest of Peninsular Malaysia. In : APPANAH,
S., NG,
F. S. and ISMAIL,R. (eds.) Malaysian forestry and forest
products research, pp. 16-21. Forest Research Institute
Malaysia, Kepong, Malaysia.
KLEINE,M. and HEUVELDOP,
J. 1993 A management planning
concept for sustained yield of tropical forests in Sabah,
Malaysia. For Ecol. Manage. 61: 277-297.
KLEJNE,M. 1997 The theory and application of a systems
approach to silviculrural decision-making. Forest Research
Centre, Forestry Department, Sabah, Malaysia.
39
NEWBERY,
D., STILL,M. J. and CAMPBELL,
E. J. 1992 Primary
lowland dipterocarp forest at the Danum Valley, Sabah,
Malaysia. I. Structure and family composition. Phil. Trans.
Roy. Soc.Landon 335 (1 275): 323-457.
R. E. 1995 The effect of selective logging on rainNUSSBAUM,
forest soil and the implications for recover);. PhD dissertation,
University of Exeter, Devon, UK.
PINARD,
M. A., PUTZ,F. E., TAY,J. and SULLIVAN,
T. 1995 Creating
project in Malaysia. J. For 93: 41-45.
PINARD,
M. A. and PUTZ,F. E. 1996 Retaining forest biomass by
reducing logging damage. Biotropica 28: 278-295.
M. G. and TAY,J. In press. Soil
PINARD,
M. A., BARKER,
disturbance and post-logging forest recovery on bulldozer
paths in Sabah, Malaysia. For Ecol. Manage.
PINARD,
M. A. and CROPPER,
W. P. In press. Simulated effects of
logging on carbon storage in dipterocarp forest. J. AppLEcol.
PUTZ,F. E. and PINARD,
M. A. 1993 Reduced-impact logging as
a carbon offset method. Cons.Bio1. 7: 755-757.
SABAH
FORESTRY
DEPARTMENT.
1989 Forestry in Sabah. Sandakan,
Sabah, Malaysia.
SIST, P,, NOLAN,T., BERTAULT,
J-G. and DYKSTRA,
D. l998
Harvesting intensity versus sustainability in Indonesia.
For Ecol. Manage. 108: 25 1-260.
TAY,J. l999 Economic assessment of reduced impact logging in
Sabah, Malaysia Ph.D. dissertation. University of Wales,
Bangor, UK.
UDARBE,
M. P,GLAUNER,
R., KLEINE,
M. and UEBELHOR,
K. 1994
Sustainability criteria for forest management in Sabah. I n 0
Tropical Forest Update 4: 13-17.
rainforest of Guyana. Tropenbos-Guyana Series 6. Tropenbos
Foundation, Wageningen, the Netherlands.
WALSH,
R. P. D. 1996 Drought frequency changes in Sabah and
adjacent parts of northern Borneo since the late nineteenth
century and possible implications for tropical rain forest
dynamics. J. Trop. Ecol. 12: 385-408.
40
Reduced impact logging as part of the domestication of
neotropical rainforest
N. R. DE GRAAF
Sub-Depatfment of Forestry, Box 342, 6700 AH Wageningen, The Netherlands.
[email protected]
SUMMARY
Reduced impact logging is only part of the process of bringing the neotropical rain forest into use. To attain sustainable yield and
economically viable production, the forest structure and composition has to be changed, a process which may be called domestication. A
selection forest structure is highly desirable for management and appears to be attainable in the near future by transformation of the forest
through an adaptation of the CELOS Silvicultural System. Results from CELOS research in Suriname since 1967 are discussed, along
with an example of the management and silviculture of a commercially based and FSC-certified timber company in the Amazon region.
Keywords: CELOS, domestication, reduced impact logging, rainforest, selection forest
INTRODUCTION
Reduced impact logging (RIL) is these days often presented
as the great solution for management of the still remaining
tropical rain forests, replacing the customary timber mining.
However, while reduced impact logging (see e.g. Hendrison
1990) is an important tool in forestry, in itself it will not lead
to sustained yield of forest land. Permanent and sustained
forest use requires forest management which determines the
forest structure as a goal to be attained. This is what
ultimately determines the system of cutting and procedures
for harvesting. Applying RIL for production of marketable
timber without any further steering of forest development
will probably result in a steady decrease in standing volumes
of timber of marketable species. The forest will increasingly
be dominated by currently non-marketable species. This was
the outcome found in long-term monitored CELOS plots in
Suriname (de Graaf 1986, de Graaf et al. 1999).
The forestry literature abounds with such warnings and
information about depletion by mere harvesting (Dawkins
and Philip 1998). An obvious remedy for this degradation is
silvicultural treatment to change the balance among species.
This has to be done with a concept in mind of what the forest
should look like. This forms part of the so-called
domestication of forest, which is a term for the adaptation of
wild forest to human needs and goals (Lamprecht 1993).
Domestication must lead to the situation that once wild and
low production forest can achieve a more highly valued
place in society as a producer of goods and services. Of
course, protected forests (nature preservation areas)
constitute a separate category of forest required for
conservation needs and not subject to domestication.
This paper offers a view on an approach to domestication
of neotropical rain forest (NRF). Reduced (or low) impact
logging will play a crucial role in this. Research work by the
Centrum voor Landbouwkundig Onderzoek in Suriname
(Centre for Agricultural Research in Surinam, CELOS), and
recent experiences in Amazonas of the Mil Madeireira
enterprise (De Camino Veloso 1998) are used as examples.
THE CELOS SILVICULTURAL SYSTEM
The CELOS Silvicultural System (CSS) as proposed by De
Graaf (1986) is a cost-effective way of growing more
marketable wood in once exploited neotropical rainforest
worked on relatively short (20-30 years) felling cycles. The
CELOS concept calls for a highly controlled system and
level of harvesting, usually not involving more than 30 m"
ha-', followed by a quite drastic release of valuable trees in
the remaining stand, effected by applying a refinement. This
has to be repeated three times, with diminishing intensity.
over the first felling cycle. Forest to be managed with the
CELOS Silvicultural System should preferably be, or have
been, exploited under the CELOS Harvesting System, which
is in fact a RIL system. The combination of both systems is
called the CELOS Management System, CMS (Hendrison
1990).
The cost per m' of standing roundwood produced is in
the order of half a man-day plus some arboricide mixture
(De Graaf 1986). The refinement treatments have been
considered to be quite drastic, not to say harsh, by some
critics. Evidence from the CELOS experiments was that a
heavy single refinement did indeed create a period of much
higher increment. After a period of about 8 years, the
competition had increased so much that a second treatment
RIL as part of rainforest domestication
became desirable. Total basal area at that stage had not
nearly climbed to the pre-logging value of about 31 mZ
ha". Even after twenty years total basal area did not exceed
26 m* ha-'. Probably any further increase in total basal area
above this value will only result from increased numbers of
large trees, still largely absent in the experimental plots (de
Graaf et al. 1999).
In the CELOS Silvicultural System it is assumed that the
old Dawkins' rule of a target of 113 - 213 of maximum basal
area for domesticated forest would remain valid (Dawkins
1958). With maximum basal area in primary forest being
around 30-32 m* ha-', the management target would be 1121 m2formaximum production. Attempts to maintain much
higher basal areas impede the regeneration of the forest; the
large trees being too dominant, and causing suppression of
seedlings and saplings. The results are very slow dynamics,
as in primary forest, only driven by few gaps per ha per year.
This type of stand appears very attractive to admirers of
majestic forests, but growth and yield are low, and much
capital is locked up in the unproductively high standing
volume, often in old and defective trees.
Species composition will also change as a result of the
silvicultural treatments applied in the CELOS experiments.
The shift depends greatly on the desirable timber tree
species decided in the Management Plan. The list should be
based not on actual marketing information, but on wood
technological research, and should be as long as possible.
Naturally, future marketability should not be over-rated for
the lesser known species with somewhat undesirable wood
technological characteristics.In 1986 the CELOS list already
had about 40 species, and in the Mil Madeireira Management
Plan more than 50 are listed (De Camino Veloso 1998).This
represents a quite high percentage of the total tree
population, as lesser known species are often scarce. An
important consideration centres on the possible elimination
of the large, dominant trees of species that are not thought
likely to be marketable timber species in the next felling
cycles. If these species play an important role in the forest
ecosystem, then reducing their dominant status might
damage the functioning of the ecosystem. The case often
cited is that of the fig trees, important as food source in times
of scarcity for, among others, vulnerable animal species
(Terborg 1992). If indeed these species are key species, then
they should be so listed with the requirement that a certain
lower limit of numbers per ha should be maintained.
The author adheres to the view of Lamprecht (1993) that
it is possible to change the structure and composition of NRF
to a certain extent without endangering the stability of the
ecosystem. Much of the attention paid to competition has so
far been concentrated on above-ground factors, but belowground factors seem likely to be important too. Combinations
of species may well be greatly influenced by the symbionts
on the roots being compatible or not with neighbouring tree
species (see, e.g., Perry 1994).
In the process of domestication the slow pace of NRF
growth has to be accelerated when economically viable
results are desirable. This has to be done carefully, as the
warnings of forest ecologists such as Ter Steege and
41
Hammond (1996) against 'speeding' indicate. Forest experts
well acquainted with the NRF have noticed that in the
Guyanas the untouched dryland forests consist largely of
highly competitive and long-lived tree species, often with
durable wood and relatively slow growth. But an increase in
production of useful timber will be necessary to enable the
forest to become a resource valued and protected by society.
No speculative extrapolations of the future structure of
the forests under CSS were made in early presentations (de
Graaf 1986), the whole emphasis being based on reporting
observations made in treated plots. From experiences over
a thirty year period with refinements and liberation
schedules in the CELOS plots in Mapane, Suriname, it has
been found that steering the forest towards a certain
desirable composition and structure is quite possible.
Whether this conclusion will hold on the long run is,
however, not yet known. Nevertheless, some predictions
seem possible. In further development of the CELOS
Silvicultural System, the first refinement might be replaced
by a marked liberation of individual Potential Crop Trees
(PCTs), as advocated by Hutchinson (1987). If regeneration
is sufficiently stimulated, the seedling recruitment dynamics
might be enough to keep up with the pace of the PCTs, and
the tree size classes intermediate between these two extreme
classes might also profit. There is as yet no clear
experimental evidence in CELOS experiments as to whether
such PCT liberation of about 20-30 PCTs per ha is as
effective in promoting seedling establishment and growth as
a strong refinement would be. In any event, liberation should
be heavy to be effective. To reduce competition effectively,
a reduction of basal area near the tree to about 15 m' ha-'
(half the untouched forest's basal area) seems advisable. At
Mil Madeireira it is intended that will be done by killing all
trees of undesirable species above 30 cm dbh over a radius
of ten metres around the PCT. It was found in CELOS
experiments that any lighter intervention produced an
unimpressive effect (de Graaf et al. 1999).
Further development of the PCTs into dominant trees
with large crowns will make the upper forest storey an open
layer with a crown cover of some 30-40 percent. This is
estimated from the potential crown sizes which are thought
likely to average 80-120 m* in vertical projection. Such
'emergent' trees do not have the size of the original large and
often very old ones in untouched forest, and with 20-30
PCTs of such final size the cover in the upper storey would
be 20-30 % or somewhat more. The storey developing below
this upper storey will be thinned out by a second treatment.
The lowest tree storey (up to 15 cm dbh) should be left alone,
as working in this would mean an intensive and expensive
type of silviculture. These poles and saplings have an
uncertain future, and are not worth much investment.
Such a forest structure would very much resemble a
selection forest. Experimental CELOS plots already show
such structure in places. Once a selection forest structure has
been reached, the balance is relatively easy to maintain by
individual tree selection (Schiitz 1989).Trees to be removed
which are below timber harvesting size may be killed
standing by poison girdling.
42
N.R. de Graaf
Studies in the experimental CELOS plots since 1967
indicate that liana proliferation is probably a small risk in
this system when canopy opening is well distributed, with a
few big gaps to be accepted where liberation oversteps the
mark. Lianas in the young stage seem to need more light than
the tree saplings amongst which they grow and thus are often
not competitive.
FOREST MANAGEMENT O F MIL MADEIREIRA
CELOS concepts have not been used to start natural forest
management in Suriname yet, but elsewhere, in the Brazilian
State of Amazonas, a forest enterprise has been set up
recently which uses the CELOS results. This is Mil
Madeireira Itacoatiara Ltda., an FSC-certified daughter
company of Precious Woods.
The site
The forest area of the commercial enterprise Mil Madeireira
Itacoatiara Ltda amounting to some 80,000 ha lies in the
vicinity of Manaus, the capital of the Amazonas State, on
the dissected plain of the 'planalto' formed on the huge and
deep Amazon basin deposits. Soils are very poor whitish
kaolinitic clays with a very high clay content (sometimes
>80%) and a high (often >90%) aluminium saturation,
coupled with a quite low CEC of 1.5-2. Drainage under
undisturbed forest is reasonable, but the soil is easily
compacted. The terrain is dissected, with sometimes steep
slopes at the edges of the plateaux. Rainfall is about 2200
mm annually, with a dry season of four months.
The socio-economic background
Only a handful of tree species at M11 Madeireira produce
wood that can be currently sold on international markets,
and even for the local market the prospects for many
potential (lesser known) timber species are bleak. Wood
technological qualities such as hardness, difficult seasoning,
difficult working, problems with painting, etc. are on the
whole not favourable. The certified export sawnwood and
pilings from Mil find a special market in Europe however.
Producing finished products such as houses, using certified
but not specified lumber, may form one of the better
strategies for improving financial results of the enterprise.
This will allow use of many lesser known species, giving
them a permanent place on the list of desirable species At
present, roundwood prices in Amazonas are low, which is
usual in a situation of apparent superabundance of the
resource. The stumpage price is traditionally neglected in
such conditions of apparent abundance. Log price at the
sawmill is then the sum of harvesting and transport costs
only. Mil Madeireira management considers that the price
of standing timber required to complete the cycle and reach
s u s t a i w yield is a b m 10-20 US$ per m3, based on the
costs of the silvicultural and other inputs needed. For Mil
Madeireira log price at the mill, the sum of stumpage,
harvesting and road transport costs, is about 40-50 US$ per
m3 roundwood (1999).The sawmill is located inside the
forest area, which saves much transport cost. The actual log
cost is comparable to that in traditional timber mining.
In Amazonas, and in fact in many regions in Brazil,
forest land can be bought at relatively low prices. Big land
owners' rights are strong, and squatters can be expelled from
their land. On the other hand, forest land ownership, as
contrasted with concession holding, offers good
opportunities to arrange ecologically sound forest
management and to provide permanent jobs for local people.
Forest resource management in the Amazon - a forest
area of several hundreds of millions of hectares -is still in
its infancy, but huge areas are already being secured by
foreign exploitation companies. Brazilian laws are regularly
adapted to new developments in Brazil, but their application
on the ground is not easy. The example set by Mil Madeireira
of Forest Stewardship Council-certified forest management
is of great importance both to the Forest Service (IBAMA).
which gives great emphasis to environment and nature
conservation, and to the logging companies that have to
improve their systems. Mil Madeireira is thus a sort of test
case.
Silviculture a n d logging
The silvicultural system set out in the Mil Madeireira Forest
Management Plan is the CELOS Silvicultural System (de
Graaf 1986), but adapted to the local situation. It is a more
advanced version, as more information, including that from
other forest research institutes such as INPA and EMBRAPM
CPATU, has become available since the system was
formulated in 1986 from results drawn from research in the
Mapane forest of Suriname.
Key characteristics for silviculture are that the forest in
the Mil Madeireira area is highly mixed and unevenaged, as
usual for NRF, with a large share of tree families - and even
species - that are also common in the Guyanas, such as
Ocotea and Nectandra spp, Manilkara spp, Hyrnenolobium
spp, and of course many Lecythidaceae. Total basal area to
a dbh of 5 cm is about 30 m? ha-', so this might be regarded
as a meagre forest. Few trees reach diameters above 1 metre.
Many such large trees are hollow or rotten, and this makes
good saw logs even scarcer than big trees are already. A
maximum of 30 m' ha-l to be harvested was set by Mil to
avoid heavy logging damage, this value being derived from
formulae for logging damage developed by Brils and Laan
in Guyana in the Iwokrama region under the guidance of Van
der Hout (1999).
Care is taken in the selection of the individual trees to be
harvested. For this selection the highly computerised system
of CMS, in which trees are numbered in the field, listed and
located on maps, works well. The computerised tree
registration system is also used for silvicultural purposes.
The selection rules explicitly require saving a part from the
harvestable trees from being taken. This is to help conserve
the stock of species in greatest demand. Annual harvested
area was planned to be around 2000 ha net, on a 25 year
RIL as part of rainforest domestication
cutting cycle, but developments in the industry and the
market will make a new plan necessary in the near future. At
least two liberation treatments are planned during the cycle,
in order to reach a sufficiently high annual increment
percentage (the increment is netted of mortality, which is
estimated at 2%). The next harvest will be of somewhat
smaller dimensions, and with slightly different proportions
of species. A local experiment on treatment has been set up
at Mil in order to be able to estimate volume increment over
a few years. Ideally, the tree spotter should have clear rules
for selection of PCTs which stem from a silvicultural
concept. In the CMS variant used at Mil, the PCTs are
actually not selected during harvest planning, but they are
registered from 40 cm dbh upwards, and their standing
volume is supplemented by larger individuals of marketable
species left standing during harvesting.
The procedures for harvesting allow the felling crew
some liberty to make changes in the list of prescribed
harvest, for example because of unforeseen defects and
other problems, but choice paths are pre-defined. The
harvest consists of trees preferably larger than 60 cm dbh, or
even more for some species with a low recovery in the mill
arising from a large sapwood zone and frequent central core
defects. Accordingly the PCTs are not interesting for the
harvest at Mil Madeireira. But this is not so everywhere. And
it could also change in the management of Mil Madeireira
in future.
The impact of the harvesting on the forest of Mil is much
less than with more traditional logging enterprises elsewhere
in the Brazilian Amazon region, and this is mainly because
of the restricted volumes taken, the application of directional
felling, and the use of relatively light machines (D4 for
winching from stump to skidding trail, wheeled skidders for
long distance skidding on trails) and harvesting under
careful planning. Most of the logs are winched towards the
semi-permanent skidding trails. Truck roads are constructed
carefully to preserve the environment, and indeed look
good, much unlike the usual 'battlefield views' loggers think
normal. It is estimated that about 5% of the forest area is
influenced, mainly by compaction, by machines, mostly on
permanent infrastructure (skidding trails, log landings, truck
roads). In this regard it is important to note that the terrain
is relatively easy, with many plateau areas. Future enterprises
should plan for this, and not buy, acquire and enter difficult
terrain lightheartedly. Such difficult terrain has to be seen
more as Protection Forest than as productive forest.
Conditions and options for selection of forest land by future
forest enterprises have been discussed by de Graaf et
a1.(1996).The logging system was fitted into the silvicultural
setting by an expert with experience as exploitation manager
of Bruynzeel in Suriname, and who studied the CELOS
Management System thoroughly. He was able to adapt and
improve the operational directives derived from the CELOS
harvesting system. Here, the logging thus became part of the
silvicultural system, and not as usual where silviculture is
reluctantly let into the logged area to 'clean up the mess'
after the loggers have maximised their harvest.
43
Management planning at Mil Madeireira
A forest management plan is compulsory in Brazil, but
management planning in natural forest is as yet very simple
and more formal than real. The Mil management plan is one
of the few examples of a plan which prescribes clearly and
elaborately what should be done and what indeed is done.
This, and the adoption of the results of the CELOS research
provided from Suriname as a start, was the basis for the
relatively rapid and easy certification of Mil. Results from
research in the Amazonas region were helpful in formulating
the Management Plan, especially the information given by
INPA, EMBRAPA and SUDAM, as well as IBAMA. The
capital needed for Mil is from European, mainly Swiss,
investors. Local investors are not interested, as capital is
very scarce in Brazil, and interest rates are very high.
Much attention has been given to nature conservation in
the plan. This is essential in productive forests, not just to
keep critical eco-minded customers satisfied, but to keep the
forest ecosystem functioning and productive. The many
forest-bound animals, and especially the 'deep forest
species' such as spider monkeys, have no place in the
agrarian landscape. Nor have large predators: these help
balance the seed predators and seed dispersers' actions.
Because of the extreme nutrient poverty of the soil, the
product has to be restricted to timber logs, to reduce nutrient
export with the harvest of such relatively high value unit
products. Bulk products such as charcoal and pulpwood
harvesting would take much more from the ecosystem. The
production and marketing of Non-Timber Forest Products
at present does not appear very promising at Mil Madeireira,
but ample scope is left for new developments in the
Management Plan. However, the importance and economic
viability of NWFP has to be proved in advance in order to
justify inclusion in the Plan. Currently even charcoal from
wood residues at the mill is not to be a viable product.
Gathering and hunting is largely forbidden in the forests of
Mil, but cannot yet be monitored, and cannot be effectively
controlled, though hunting laws in Amazonas are strict.
CONCLUSIONS
The role of RIL in the process of designing a silvicultural
system should be to provide various options for further
sustained forest management. In the Mil Madeireira case it
has become clear that selection of the harvestable trees is in
fact a silvicultural decision. This has for long been the case
in classical forest management in Europe, as well as in
several countries in South and East Asia and in North
America. Only the clearfelling system can avoid such
silvicultural guidance, because regeneration is then usually
by planting or direct seeding. A special case is the retention
of Potential Crop Trees, which is likely to become more and
more popular in management of Humid Tropical Forest.
Here, under RIL, the logging operator has to accept the
choice made by the tree spotter, even if it appears that a large
44
N.R. de Graaf
volume of harvestable wood is left standing. This is hard
when the concession has to be left after a first harvest has
been removed, but if the next harvest belongs to the same
operator, time and regrowth play a big role. Long-term
concession tenure may help silviculture to carry out its task
of replenishing the resource.
Where forest management proceeds towards classical
selection felling in succeeding cycles, the frequency of
harvesting may be increased to 10 or even 5 years, with a
correspondingly low volume taken per ha. The principal
reason for this is that an annual mortality of 2% eliminates
too much potentially utilisable timber over 20-30 year long
cycles. Trees that show signs of reduced increment and
reduced vitality in classical selection forest are the first
choice for harvesting. The crux is that one should know the
signs of such conditions for each species. and this is not easy,
even for locally trained and experienced tree spotters. But it
must be possible to do it, since trees show such slowing of
growth not only by stem characteristics, but also by crown
habit, flowering and seed bearing and leaf characteristics,
etc. Rapid calculations will show that the lower volume
limit for economic harvesting is higher than the volume
available each 10 years (in the case of Mil Madeireira this
could be 15 m3 ha-'), but the fact remains that many
traditional loggers enter the forest to harvest only a few
cubic metres during creaming operations. One is tempted to
think that harvesting cost may be not felt to be so punishing
as suggested by textbooks.
Such frequent harvesting cannot be carried out by
conventional logging, with bulldozers entering the stand and
driving up to the stump, as this would damage much
undergrowth and smaller trees, and would compact too
much soil. In forest with modest sized logs, complete
winching as done by Mil will be the solution, as the machine
then stays on the semi-permanent skidding trail. Damage
then is mainly due to felling. This has to be controlled by
directional felling, avoiding not only hang-ups and positions
that are wrong for winching, but also avoiding hitting PCTs.
Winching does very little damage when carried out expertly
by a trained crew, as is the case at Mil Madeireira. Most of
this is already known and used as techniques in European
selection forests for many years.
I conclude that RIL will play an essential role in the road
towards selection forest management in the neotropical rain
forest.
REFERENCES
DAWKINS,
H.C. 1958 The management of natural tropical highforest with special reference to Uganda. Imperial Forestry
Institute, University of Oxford. 155 pp.
DAWKINS,
H.C. and PHILIP,M.S. 1998 Tropical moist forest
silvic~ltureand management: a history of success andfailure.
CAB International, Wallingford, U.K. 359 pp.
DE CAWNO
VELOSO,
R. 1998 Research needs for a low impact
logging operation in the Brazilian Amazon: the case of
Precious WoodslMil Madeireira Itacoatiara. In: Seminar
proceedings Research in tropical rain forests: its challenges
for the ficture. The Tropenbos Foundation, Wageningen, The
Netherlands, pp. 43-58.
DE GRAAF,N.R. 1986 A silvicultural system for natural
regeneration of tropical rain forest in Suriname. Dissertation
Wageningen Agricultural University. 250 pp.
N.R., KONING,
DE GRAAF,
D.and SPIERINGS,
M. 1996 Some
conditions and possibilities for successful application of the
CELOS Management System. BOS NiEuWSLETTER, V0115
(2), NO 34, pp 74-83.
DEGRAAF,
N.R., POELS,
R.L.H. and VAN ROMPAEY,
R.S.A.R. 1999
Effect of silvicultural treatment on growth and mortality of
rainforest in Surinam, over long periods. For. Ecol. and
Manage. 124: 123-135.
HENDRISON,
J. l990 Damage-controlled logging in managed
tropical rain forest in Suriname. Dissertation Wageningen
Agricultural University. 204 pp.
HUTCHINSON,
1.D. 1987 Improvement thinning in natural tropical
forest. In: MERGEN,
F. and VINCENT,
J.R. (eds.) Natural
management of tropical moist forests Yale University, New
Haven.
LAMPRECHT,
H. 1993 Silviculture in the tropical Natural Forests.
In: PANCEL,
L. (ed.) Tropicalforestry handbook, Vol 1 and 2.
Springer Verlag, Germany, pp. 727-810.
PERRY,
D.A. 1994 Forest ecosystems. The John Hopkins
University Press, Baltimore, London. 649 pp.
SCHUTZ,
J-P,, 1989 Der Plenterbetrieb. Fachbereich Waldbau,
ETH, Ziirich. 54 pp.
TERBORG,
J. 1992 Diversity and the tropical rain forest. Scientific
American Library, New York. 242 pp.
TERSTEEGE,
H. and HAMMOND,
D.S. 1996 Forest management
in the Guianas: ecological and evolutionary constraints on
timber production. BOS NiEuWSLETTER Vol 15 (2) 110.34:
62-69.
rain forest of Guyana. Dissertation University of Utrecht,
Tropenbos Series 6, Tropenbos-Guyana Programme,
Georgetown. 335 pp.
45
Benefits, bottlenecks and uncertainties in the pantropical
implementation of reduced impact logging techniques
DAVlD S. HAMMOND l, PETER VAN DER HOUT *, RODERICK J. ZAGT 2, GODFREY MARSHALL 3,
JULIAN EVANS 3and DAVlD S. CASSELLS
lwokrama International Centre for Rain Forest Conservation & Development, 67 Be1 Air, PO Box 10630, Be1 Air,
Georgetown, Guyana
Tropenbos-Guyana Programme, 12e Garnett St., Campbellville, Georgetown, Guyana
Guyana Forestry Commission, Kingston, Georgetown, Guyana
SUMMARY
Reduced impact logging has been shown to be environmentally beneficial by reducing damage to the forest stand and soils, but is only
one component of good forest management. The implementation of RIL is largely contingent on satisfying concerns about cost to the
producer and the values of benefits. While many of the direct costs and benefits associated with RIL have been quantified, there are still
a number of unanswered questions and potential bottlenecks. The cost of training, extra wage demands, monitoring, verification and
foregone timber have not always been accounted for in cost comparisons between RIL and conventional logging systems. While direct
benefits of employing RIL through waste reduction could be considered universal, benefits derived through other income generation
schemes, such as carbon offset, are not so clear. Institutional and economic constraints will continue to reduce the likelihood of RIL
implementation in those regions where lease conditions promote short-term management objectives.
Keywords: damage, costs, benefits, tropical forests, certification.
INTRODUCTION
Reduced impact logging (RIL) techniques, like many other
initiatives to improve the way in which we manage tropical
forests, have been developed and refined through decades of
experience under a wide array of biophysical and socioeconomic conditions and management objectives (Nicholson
1958, Mattson MBrn and Jonkers 1981, Uhl and Viera 1989,
Hendrison 1990, Pinard and Putz 1996). Literature documenting the effects of adopting RIL techniques has grown
tremendously, in particular over the last decade (see Sist,
this volume). The RIL workshop held in Georgetown,
Guyana in April 1999, brought together individuals who
had participated in many of the seminal projects looking
at the use of RIL techniques. The workshop produced a
series of papers, many of which are contained in this issue
of the International Forestry Review, and held several days
of working group discussions aimed at distilling the best
practices and lessons learned from a diverse assortment of
site case studies and professional perspectives. This paper
aims to synthesise and build upon the results of the working
group sessions while drawing heavily upon the case study
papers published in this volume and other publications that
have been produced by contributing members of the Guyana
RIL workshop. The workshop participants also identified
where the interface between RIL techniques and forest
management objectives remains clouded by insufficient
information and where potential bottlenecks to the successful implementation of RIL are most likely to occur. These
are summarised here as a series of key factors influencing
the beneficial implementation of RIL in tropical forests.
BENEFITS AND COSTS OF RIL IMPLEMENTATION
The working group discussions and papers presented at the
Georgetown RIL workshop repeatedly focused on a set of
interlinked and often competing objectives associated with
the use of RIL techniques, viz:
to minimise changes in forest structure
to maintain a growing stock of marketable timber trees
to optimise the harvesting cycle
to limit the impact on other non-timber forest goods and
services
to maximise the carbon sink role of tropical forests
to reduce inefficiencies in harvesting to the financial
benefit of producers
to contribute to a change in the growing market perception that the cost of tropical timber production is
unacceptably high when environmental costs are considered.
46
D.S. Hamrnond et al.
All of these objectives are directly or indirectly associated with the main operational advantage of employing RIL
techniques: a reduction in damage during the harvesting
process. It was repeatedly emphasised, however, that RIL
techniques are only one, albeit vital, component of good
forest management. Sustainability will not be achieved
through the adoption of RIL alone, when other aspects of
forest management (e.g. harvesting intensities, post-harvest
silvicultural control, see de Graaf, this volume, pp. 38-42)
are not considered in the same framework. Nonetheless,
RIL was considered to be an essential precondition of good
stewardship and the possible market benefits derived from
achieving such a classification.
A number of operational elements were identified which
appeared to consistently contribute to achieving the objectives of employing RIL techniques: see Jonkers and van
Leersum (this volume pp. 11-16), Sist (pp. 3-10), van der Hout
(24-32) and Table l . They outline best practice because they
delimit the extent to which techniques can be standardised
and still meet most of the expected objectives. However, as
shown in Table 1 a consensus has not always been reached
when considering whether or not benefits outweigh the costs
of implementing RIL methods. Operational elements of RIL
which are repeatedly characterised by low cost and high
benefit would be most important in defining best practice.
FACTORS AFFECTING NET BENEFITS O F RIL
Working group discussions at the workshop often reinforced
many of the rapidly changing perceptions of the universal
applicability of RIL. Many of these perceptions focus on the
limits to applicability of standard RIL techniques when
biophysical and socio-economic conditions surrounding
timber production vary widely (see Sist, this volume). While
most studies suggest that there is generally an overall net
benefit in using RIL techniques, a number of uncertainties
and potential bottlenecks remain, several of which are
highlighted below.
Harvesting intensities
Sist et al. (1998) have questioned the ability of RIL guidelines, as defined, to deliver on their promise in SE Asian
forests where the felling intensity IS often 2 to 8 times higher
than in forests either in Africa or South America (also see
Sist, this volume). Van der Hout (1 999) suggests further that
the 'intensity clause', defined as the loss of advantage in
employing RIL when harvesting reaches a certain intensity,
is equally applicable to forests in South America where the
range of felling intensities normally occurs at the lower end
of those which typify operations in SE Asia. The reason that
the 'intensity clause' applies in such cases is that felling
intensities can be just as high, due to an aggregated
distribution of commercial stems. Such clumping is one of
the majm constraints to achieving a felling intensity which
will optimise benefits when confronted with the decision to
reduce damage and, in the process, forego harvestable
timber (see Sist, Pinard et al., this volume pp. 33-39, van
der Hout 1999). Defining the allowable felling intensity
would appear to be the first step in determining the
applicability of RIL in any forest area, a point that was
largely agreed upon by the working groups at the
Georgetown RIL workshop.
Conclusion
The use of RIL techniques will ultimately fail to produce the
expected benefits when harvesting levels overshadow the
gains made through carefulplanning. Harvesting levels are
highly sensitive to the spatial distribution of conmercial
stems and the ecological and economic implications their
distribution has for reducing damage at the expense of
foregone tmbel:
Forest carbon storage, productivity, fire a n d climate
change
RIL prescriptions were employed in SE Sabah in order to
evaluate their potential role as a carbon-offset method
(Pinard and Putz 1993, Pinard et al. this volume). The
authors estimated that the use of RIL techniques could retain
an additional 36 tonnes of carbon per ha over that remaining
after conventional harvesting within two years of logging
(Pinard and Putz 1993), mainly due to a reduction in
necromass associated with mortality among damaged stems
(Pinard and Putz 1996). While substantial carbon offset
benefits can accrue through the use of RIL in SE Asian
forests, a first assessment of the potential for carbon offset
in Guyanan forests suggests that carbon benefits derived
from reducing damage could be quite small (Ter Steege
1998). Though estimates of carbon offset in tropical forests
are plagued by a very patchy understanding of the size and
predictability in time and space of carbon stores within
tropical forests, the considerable variation in average stem
size and height (see Sist, this volume), wood density (see
Hammond et al. in press), rainfall and stand turnover (see
Philips et al. 1994) between different forest regions suggests
that carbon offset benefits accrued through the adoption of
RIL techniques are likely to vary considerably.
RIL may reduce the likelihood of forest fire by decreasing available fuel loads relative to uncontrolled harvesting
regimes (Holdsworth and Uhl 1997), but this benefit will
vary depending on the severity of rainfall failure and the
moisture-conserving capacity of forests during El Niiio
periods (Goldammer and Seibert 1990, Hammond and Ter
Steege 1998).
Conclusion
Carbon offset benefits accrued through the use of RIL
techniques are strongly influenced by stand productivitj,
structure, and variation in those forest attributes, such as
wood density, soil type atld hydrological conditions, which
influence the dynamics of forest carbon stores (stems, soil,
litter, etc.).
Benefits, bottlenecks and uncertainties
47
TABLE1 Key operational elements identijied at the RIL workshop and in references as characterising 'best practice' in RIL and the
level of support concerning the costs and benefits attached to their use
0 = contradicting references, +-+++ = weak to strong support based on 2-3,4-5 or >5 references (including workshop outcomes) in
agreement. Note benefits and costs are not paired by row or given in order of importance. References refer to both benefits and costs
in same row. Some references used in assessment are nested within those presented here and are not listed separately. Full reference
assigned to each number denoted by (number) in bibliography.
-
Key Operational Element
Training
Benefits
Support
training courses and
workshops
higher wage demands
higher staff turnover
reduced unit cost of harvesting
reduce damage to PCTs
higher quality timber
fewer accidents and less
equipment damage
better compliance with guidelines
Pre-harvest inventory
trees to be harvested
potential crop trees
(PCTs)
NTFP trees
protected trees
Costs
optimise immediate timber
revenue
optimise future timber stock
inventory and planning
inventory and planning
discounted future price
foregone timber revenue
other sources of revenue
biodiversity conservation
better relations with local residents
contribution to forest function
foregone timber revenue
promote competing noncommercials
keep open option values
meet nat'l and int'l standards
Climber cutting
feller safety
improved recovery
improved future yields
field crew and equipment
lost plant diversity
reduced wildlife resources
Topographic assessment
efficient access
GIS, cartographic staff &
equipment
improved environmental protection
Protected areas
delimitation
buffer zones
conservation areas
sacred areas
protect forest hydrological
function
conserve sensitive aquatic life
conserve genetic diversity
equipment
equipment
strategic field inventory
crew & equipment
provide refuge for sensitive
species
retain examples of pre-logged
forest
better relations with local
residents
contribute to national cultural
heritage
conserve forest biodiversity
Extraction network planning
road and skidding
drainage
trail alignment
efficient access to harvestable trees
retain PCTs
landinglyard/market minimise environmental impact
placement
good access conditions
skidding trail
ensure efficient access to felled
marking
trees
minimise damage
Archaeology/anthropology
services
+++
+++
+++
++
++
+++
+++
equipment
equipment
field crew & equipment
Support
Reference
48
D.S. Hammond et al.
TABLEl (continued)
Key Operational Element
Felling
control
direction
Benefits
improve recovery
reduce accidents
reduce losses to PCTs
Support
Costs
felling crew & good
quality equipment
training
felling accessories (wedges,
sledges, etc)
reduce canopy openness
enhance skidding efficiency
improve recovery
Winching
long winch distance
Skidding
low blade use
reduce soil and stand damage
field crew & equipment
reduce skidder fuel use
training
winch cables and
accessories
reduce soil damage
avoid steep slopes
reduce soil and stand damage
minimise distance
reduce equipment damage
reduce accidents
reduce operating time and fuel use
foregone timber
Forest closure
skidding trail cross- reduce gully erosion
drains
secondary road
reduce poaching
closure
clean up non-organic minimises pollution hazards
waste
construction and
maintenance
construction and
maintenance
collection and disposal
Monitoring
operational costs
field crews and equipment
reduce costs through efficiency
improvements
soil and stand damage compliance with nat'l and
int'l standards
correct errors and highlight
successes
growth and
assess yield expectations
recruitment
identify saleable properties of
next harvest
opportunities to ameliorate
impact on wildlife
damaging practices
maximise wildlife support services
to timber trees
alter perception of poor
stewardship
Verification
compliance with
Codes-of-Practice
compliance with
international
standards
avoid financial penalties
avoid civil court
retain right to harvest timber
(on public lands)
alter perception of poor
stewardship
greater access to restricted markets
price premium on certified
products
computers and software
forestry services
botanical services
wildlife biology services
publicity and publications
internal auditing
report production
independent certifier fees
registration fees
Support
Reference
Mitigation of fire risk through a reduction in damage
only works i f the severity of fire-driving events is of
moderate duration and intensity. Reduced damage will still
heighten susceptibility to fire over unlogged forest under
extreme scenarios. Climate change may increase fire risk in
some loggedforests and reduce or negate any carbon offset
benefits accrued over shorter periods.
Regeneration strategies for target commercial species
Many major timber species, such as mahogany (Swietenia
macrophylla), duabanga (Duabanga moluccuna), white
seraya (Parashorea spp.), goupi (Goupia glabra), cedro
(Cedrela odorata), azobB (Lophira alata) and balsa
(Ochroma pyramidale) require considerable canopy disturbance to regenerate, often through catastrophic events such
as fire, flooding or hurricanes, while others, such as
greenheart (Chlorocardium rodiei), rosewood (Dalbergia
spp.), iroko (Chlorophora excelsa) and ironwood
(Eusideroxylon spp.) largely establish best under lower
levels of canopy fragmentation, where competition from fast
growers is reduced (e.g. Foster 1990, Gullison et al. 1996,
Whitmore and Brown 1996). An aim of RIL is to ensure that
logged forest remains stocked with marketable trees, but
where regeneration of the main marketable species in the
stand depends on significant breaking up of the canopy, the
role of RIL techniques within the overall forest management
strategy would need to be carefully assessed.
Conclusion
Stable supplies of currently-marketed timber species will
only be enhanced in stands where RIL techniques are
employed if the resultant level of disturbance is consistent
with the regeneration requirements of the target timber
species. Alternatively, target species and market focus will
have to change or some form of silvicultural treatment will
be needed to alter conditions after hawesting.
Soils and topography
Participants at the RIL workshop concurred that soil compaction and loss of hydrological function during log extraction were two of the most obvious adverse impacts that
poorly planned harvesting can impose on the forest ecosystem.
While it is generally clear that pre-harvest planning of
skidding trails will reduce damage to soil and stand compared with unplanned methods, the relative contribution of
detailed skidding trail management to the overall benefits
accruing through the use of RIL practices is likely to vary
from site to site. Rubber-tyred skidders are not capable of
working steep slopes, such as characterise many forests in
SE Asia, and the consequent levels of soil disturbance
caused by crawler tractors, even under RIL, may still be
considerable relative to unlogged forests. The same applies
to large tree dimensions characteristic of logging in SE Asia
49
and central Africa. In contrast, gently undulating, sandy soil
terrain stocked with relatively small stems in parts of South
America and Africa is more easily managed with rubber tyre
skidders with minimal soil disturbance.
The environmental benefits from using RIL skidding
techniques will depend largely on the magnitude of existing
background levels of disturbance. For example, waterways
which are normally not subject to large influxes of sediment
are likely to undergo more drastic biogeochemical change
than those which, through their proximity to steep, exposed
slopes are continuously fed large volumes of sediment.
Furthermore, this impact may not be proportional to the
increase in sediment inputs caused by harvesting, particularly where biota have adapted to an unusually low suspended sediment environment (Junk 1984).
Conclusion
Benefits derived from a reduction in soil disturbance
and erosion will vary depending on the prevailing background conditions. Differences in topographical, soil and
hydrological conditions will alter the savings made by
RIL when the costs of these environmental impacts are
counted.
The interaction between slope, soil and stem size will
constrain the achievable environmental benefits without
further technological and financial inputs (e.g. aerial
extraction). Damage associated with the type of machinery
used to extract timbel; not the way in which it used, is
limiting when conditions prevent the use of less destructive
techniques.
Perceptions within the tropical timber market
Most workshop participants agreed that distorted perceptions of tropical timber production and RIL plagued progress
in persuading producers to adopt better management practices (including RIL techniques) and convincing consumers
that tropical timber harvesting is not necessarily always
degrading. The working groups confirmed that distorted
perceptions were one of the most serious constraints to
progress in achieving good forest management. Without an
objective perspective, the linkage between investing in and
implementing sound management practices and the confident consumption of sustainably produced timber products
is difficult to establish.
Producer concerns with RIL techniques often revolve
around the perception that implementation costs will be
greater than those currently incurred using less detailed
planning procedures. In some cases, this perception may
seem justified where (mainly pre-harvest planning) costs are
viewed without considering all of the benefits accrued by
implementing RIL methods. A higher volume recovery,
however, is alone sufficient to make the implementation of
RIL techniques a cost-neutral or cost-reducing exercise in
most cases (Montenegro 1997, Baneto et al. 1998, Holmes
et al. 1999, van der Hout 1999). In other regions where the
50
D.S.Hammond et al.
costs may be higher, particularly when considering the
opportunity costs of foregoing considerable volumes of
marketable timber, other spin-off benefits, such as carbon
offset, could potentially absorb the additional cost of
implementation (e.g. Pinard et al. this volume).
A comprehensive set of case studies and well-considered
analyses are needed to document the full costs and benefits
of implementing RIL and these results need to be disseminated widely to all the major stakeholders in the timber
industry.
Conclusion
Negative perceptions on behalf of both producer and
consumer willprevent a change in the way in which tropical
forests are managed and the likelihood of RIL implementation. RIL research and development will only be validated
ifperceptions are fostered through a balanced, and honest,
appraisal of the costs and benefits that the private sector
can expect to experience by implementing RIL.
Conclusion
Verifiing the effects on costs of RIL techniques will need to
account for the varying cost of training and the subsequent
rise in wages. Higher wage demands than might otherwise
arise may reduce the total benefits received through implementation ofRlLpractices, but this can be seen as a medium
to long-term investment. The unit cost of this one-time
investment should be recouped over a specific term of
production, unless job turnover rate or wage demand
frequency is high enough to limit unit cost reduction (see
Heinrich 1996, Pinard et al. this volume). The cost of
training as a precursor to increased wage demands needs
to be assessed.
Without well-trained, motivated and satisfactorilj paid
field crews, the likelihood of achieving the objectives of RIL
is extremely low. The role of externalfinancial support in the
training of logging crews to reduce employer risk may be
crucial for implementation.
Information management
Required technical competence
Few studies have adequately addressed the costs and
benefits of training, although training is by virtue a necessary prerequisite to the implementation of RIL practices (see
Dykstra and Heinrich 1996, Pinard et al. 1995, Armstrong
this volume pp. 15-21). Workshop participants identified
inadequate training and lack of skilled personnel as major
obstacles to the successful implementation of RIL.
However, while the need for training is fundamental,
many studies appear not to have incorporated the cost of RIL
skills development into the final assessment of net benefit
(e.g. Johns et al. 1996, Barreto et al. 1998, van der Hout
1999). Holmes et al. (1999) and Barreto et al. (1998)
assumed that salaries for the main technical staff would
remain the same in their comparative analysis of costs and
benefits underlying conventional and reduced impact logging techniques. Yet Barreto et al. (1998) found when they
hypothetically doubled the salaries of RIL staff as a means
of accounting for the higher level of required skills, the
increase in profitability associated with RIL declined from
31% to 12% over unplanned logging. Sist et al. (1998)
emphasised that RIL techniques require well-trained staff
and payment schemes must account for the higher quality of
work. This too is likely toincrease the wage costs associated
with the use of RIL techniques. Winkler (1997) noted the
considerable increase in wage costs associated with a
well-trained, healthy and well-supported RIL forest crew
employed by Precious Woods in Brazil. The cost of training
and subsequent wage upgrading is also likely to vary
from site to site, depending on a host of factors which
influence wage rates, job competition, and the extent and
frequency of training required. For example, Holmes et al.
(1999) reported a cost of US$0.21 m-"or training of staff
ilr Brazil, while Montenegro (1997) cites a cost that is
mre h n double this figure (US$0.59 m-') at a site in
Ecuador.
The majority of workshop participants did not consider that
an incomplete knowledge of the forest system should act as
a constraint on good practice (cf. Hendrison 1990). Winkler
(1997), Sabogal (1998) and many others have emphasised
that scientific results need to be made available to practitioners in a form which is meaningful to them. Barreto et al.
(1998) suggest that a lack of information available to the
timber industry represents one of the main obstacles to
adoption of good forest management practices. In this
respect. the prolific output of scientific and policy papers
concerning reduced impact logging stands in sharp contrast
to the number of practitioner tailored publications (e.g.
Amaral et al. 1998). If robust mechanisms for information
transfer to industry are put in place, there is still the need for
industry to effectively manage this information for its own
use. Armstrong (this volume) suggests that operations need
to take advantage of advances in information technology to
speed up the planning process if RIL and other components
of good forest management are to be carried out successfully. Even with the use of advanced technology it is difficult
to imagine the successful implementation of RIL without a
structured system of information handling.
Conclusion
The capital cost of such items as databases, GPSs, computers, printers, GIS and data loggers associated with the
effective transfer and management of information for
improved forest management, including the use of RIL
techniques, must be incorporated into the calculation of
costs and benefits associated with better planning. Technology and information should to be transferred in a way that
is tailored to the needs of practitioners.
A more data-intensive forest management system is
vulnerable when systems of information management are
not in place. An increase in the amount of information
generated to improve forest management will not achieve
improvements i f this information is not available to industv
and regulatory agencies. And an increase in the flow of
information to industry and regulatory agencies will not
lead to improvements if this information is not effectively
managed.
Implementation of forest management plans
Though the term reduced impact logging is recent, the
guidelines and techniques are not particularly new (see, e.g.,
Gilmour 1977). Unfortunately, over the last 20 years, very
few logging operators have been willing to embrace methods designed to reduce damage, mainly because field
practitioners are by tradition conservative when it comes to
adopting new approaches (Hamilton and King 1983), and
low-impact harvesting was often not the approach traditionally employed in temperate forests (e.g. Williams 1989).
Many industrial-scale logging ventures are operated by
foreign or transnational companies who may fail to understand the need to reduce waste in order to achieve profitability when forest productivity and socio-economic conditions
at their other site holdings have made this unnecessary.
Most companies hold relatively short-term leases on public
forest lands and this provides a disincentive for companies
to invest in retaining potential crop trees (PCTs) if they do
not expect to be harvesting them in the future. The workshop
highlighted lease term as a major constraint to achieving best
practice. Combined with extremely poor rent1 capture by
governments (Repetto 1988) the incentive to carry out
unplanned 'creaming' of the most immediately saleable
fraction of the stand is compelling; the merits of developing,
implementing and monitoring a system of RIL planning are
not.
Workshop participants suggested that it is desirable that
a single concessionare in a country should take the initiative
in applying RIL at an operational scale. Documented results
from their efforts need to be broadly distributed to industry
and government and then disseminated through a far-reaching campaign of publicity. Industry and government would
then be in a better position to evaluate what kind of incentives
and regulations are needed to achieve good management.
Conclusion
Implementation of RIL guidelines depends largely on the
timber industry's 'willingness to pay'. If the costs of
implementing RIL (net of financial benefits) are at a level
acceptable to industry and traditional views become more
amenable to emerging perspectives (e.g. due to declining
market access), then RIL will be widely adopted. Macroeconomic and sectoral policies blanket industry's decisionmaking and these must promote, not erode, the incentive to
adopt good forest management practices. Mechanisms to
reduce the the financial burden borne by producers who
wish to make the transitionfrom traditional logging to RIL
should be developed, particularly where net financial
benefits are not certain.
51
Monitoring and verification of good forest management
Confirmation that forest management plans are consistent
with practice is another fundamental component of implementing RIL techniques, and assessment of harvesting
performance is central to nearly all forest Codes of Practice
(Dykstra and Heinrich 1996). Operational monitoring provides a basis on which to gauge the extent to which a set of
business practices, both in the forest and office, are contributing to profitability, yet meeting the internal standards set
out in a management plan and those external standards
developed through consultation by various national and
international agencies and regulatory bodies.
Expenditure on monitoring and verification has not
normally been incorporated into cost comparisons between
conventional logging and RIL (Barreto et al. 1998, Holmes
et al. 1999, van der Hout 1999), though data in Montenegro
(1997) suggest that post-harvest monitoring can add 1.S% to
the cost of unplanned logging.
The transfer of benefit away from producers in the form
of certifier fees would seem a strong disincentive to invest
in verifying, let alone implementing, good management
practices, unless there is perfect collusion among consumers
in their purchase of certified products. Many tropical timber
producers' earnings, particularly in South America and
Africa, can be low because prices for tropical sawnwood and
plywood are largely subject to a high elasticity of substitution in foreign markets (Barbier et al. 1994), and the chain
of end-use processing often results in only a small fraction
of the final retail price being captured by the producer
(Hammond et al. in press). Often it is the level of rent
capture and duty-free concession alone that determines
whether tropical logging can be a profitable industry or not
(Repetto 1988). The incentive to invest in monitoring and
verification can vary depending on the uncertainty of prices,
rents and tenures.
Conclusion
Monitoring and verification is a vital component of RIL, but
the costs of this have not been adequately quantified.
Certification is currently expensive, but may stimulate the
implementation of RIL and SFM by providing a gateway to
privileged timber markets. Implementing RIL practices-on
the assumption that planning is sufficient to achieve sustainable management, without incurring the cost of monitoring, assumes that management need not be 'adaptive'.
Benefits from monitoring may accrue over subsequent
harvests as lessons are learned and integrated to improve
the efficiency, and thereby savings, of the operation. Alternutively, the costs of monitoring need to be assuaged
through accredited certification bodies and procedures
developed with the client's ability to pay in mind.
l
Rent is the economic term for revenue that would accrue to the
forest owner were a fully competitive market in standing timber
to operate. Ed
52
D.S.Hammond et al.
OVERALL CONCLUSIONS
There is a growing literature relating t o the costs and
benefits of implementing RIL techniques. The Georgetown
RIL workshop clearly identified a set of operational
elements which could be considered as key attributes to a
successful RIL system (see Table 1). Most participants, and
nearly all literature sources, view these as the basic framework on which RIL is practiced. There are still many
unanswered questions, particularly with regard to the actual
cost of implementing RIL and the scope for extending
benefits by finding a market for spin-off services (such as
carbon offset) or rewarding private industry for minimising
its impact on important forest functions ( e . g . water quality
control and fire prevention) where this entails a cost
(including training and other transitional costs). The final
basket of benefits which can accrue through the use of RIL
techniques, however, will reflect regional variation in the
environmental, economic and social conditions of the day
and not all baskets will prove to be equally bountiful.
The workshop was also successful in identifying several
key economic, institutional and technical bottlenecks. Most
technical bottlenecks are surmountable but many are still
perceived to exist due to poor awareness of case study
results. Several activities to promote improved logging
techniques have been initiated ( e . g . T F F et a l . in Brazil,
Pinard et al. in Sabah) or are under way (Tropenbosl
IwokramdGuyana Forestry Commission et al. in Guyana).
This workshop showed that attention should now focus on
the identification of key mechanisms that need to be
introduced to overcome the economic and institutional
bottlenecks to the adoption of more sustainable logging
practices.
CROME,
F.H.J., MOORE,
L.A. and RICHARDS,
logging damage in upland rainforest in north Queensland. For
Ecol. Manage. 49: 1-29. (7)
DEBONIS,
J. 1986 Harvesting tropical forests in Ecuador. J. Fox 84:
43-46. (8)
DYKSTRA,
D. and HEINRICH,
R. 1992 Sustaining tropical forests
through environmentally sound harvesting practices. Unasylva
43: 9-15. (9)
DYKSTRA,
D. and HEINRICH,
R. 1996 Forest codes ofpractice. FAO
Forestry Paper 133. FAO, Rome. (10)
EK, R. 1997 Botanical diversity in the tropical rain foresr of
Guyana. Tropenbos-Guyana Series 4. Georgetown, Guyana.
(11)
FEARNSIDE,
P.M. 1989 Forest management in Amazonia: the need
for new criteria in evaluating development options. For Ecol.
Manage. 27:61-79. (12)
FORESTSTEWARDSHIP
COUNCIL 1999 Principles and criteria for
forest stewardship. Document 1.2. www.fscoaxx.org(13)
FOSTER,
R.B. 1990 Long-term change in the successional forest
community of the Rio Manu floodplain. In: GENTRY,
A.H. (ed.)
Four neotropical forests. Yale University Press, New Haven,
Ct.
GILMOUR,
D. 1977 Logging and the environment with particular
reference to soil and stream protection in tropical forest
situations. In: Guidelines for watershed management. FAO
Conservation Guide 1. FAO, Rome, pp 223-235. (14)
GOLDAMMER.
J.G. and SEIBERT,
B. l990 The impact of droughts
and forest fires on tropical lowland rain forest of East
Kalimantan. In: GOLDAMMER,
J.G. (ed.) Fire in the tropical
biota. Springer-Verlag, London, pp. 11-31.
GULLISON,
R.E., PANFIL,S.N., STROUSE,
J.J. and HUBBELL,
S.P.
1996 Ecology and management of mahogany (Swietenia
macrophylla King) in the Chimanes Forest, Beni, Bolivia. Bot.
J . Linnean Soc. 122: 9-34.
H. 1998 Propensity for fire in
HAMMOND,
D.S. and TER STEEGE,
Guianan rainforests. Cons. Biol. 12: 944-947.
HAMMOND,
D.S., ZAGT,R.J., WELCH,I. and VAN DER HOUT,P. In
press. The e c o l o g ~utilisation and management of Greenheart
and the Greenheart-bearing tropical forests of Guyana.
REFERENCES [Note: figures in bold refer to the references
included in Table l ]
AMARAL,
P,, VER~SSIMO,
A., BARRETO,
P. and VIDAL,E. 1998
Floresta para sempre; um manual para a produ@o de madeira
na AmazBnia. IMAZON, BelCm, Brazil.
APPANAH,
S. and PUTZ,F.E. 1984 Climber abundance in virgin
dipterocarp forest and the effect of pre-felling climber cutting
on logging damage. Malay. For 47: 335-342. (1)
BARBIER,
E.B., BURGESS,
J.C., BISHOP,
J. and AYLWARD,
B. 1994
The economics of the tropical timber trade. Earthscan Publications, London. (2)
BARRETO,
P., AMARAL,
P., VIDAL,
E. and UHL,C. l998 Costs and
Amazonia. FOREcol. Manage. 108: 9-26. (3)
BROTO~SWORO,
E. 1991 Indonesian forest resources and manageC. (eds.) Proceedment policy. In: HOWLER,D. and SARGENT,
ings of technical workshop ro explore options for global
forestry management. IIED, Bangkok, pp. 254-262. (4)
BRUENIG,
E. 1996 Conservation and management of tropical
rainforests: an integrated approach to sustainability. CAB
International, Oxford. (5)
C E Q E Rl.,
~FALCK,
,
J., GARCIA,
A., GOH,F. and HAGNER,
M. 1994
Rtducidg impact without reducing yield. Tropical Forestry
Update 4: 9- 10. (6)
Tropenbos Series. Wageningen, Netherlands.
HAMILTON,
L.S. and KING,P.N. 1983 Tropical forested watersheds. Westview Press, Boulder, Co. (15)
HEINRICH,
R. 1996 The FAO programme on environmentally sound
forest harvesting operations. In: DYKSTRA,
D. and HEINRICH,
R.
(eds.): Forest codes of practice. Contributing to environmentally sound forest operations. FAO Forestry Paper 133. FAO,
Rome, pp. 1-7. (16)
HENDRISON,
J. l990 Damage-controlled logging in managed rain
forest in Suriname. Ecology and management of tropical rain
forests in Suriname 4. Wageningen Agricultural University,
Wageningen, Netherlands. (17)
HOLDSWORTH,
A.R. and UHL, C. 1997 Fire in Amazonian
selectively logged rain forest and the potential for fire reduction. Ecological Applications 7: 713-725.
HOLMES,
T.P., BLATE,
G.M., ZWEEDE,
J.C., PEREIRA,
Jr., R., BARRETO,
P,. BOLTZ.
F. and BAUCH.
R. 1999 Financial costs and benefits
ofreduced-impact logging relative to conventional logging in
the Eastern Amazon. Forestry Private Enterprise Initiative
Working Paper. Tropical Forest Foundation 1 Funda~lo
Floresta Tropical / USDA Forest Service. Southeastern Center
for Forest Economics Research, Research Triangle Park, NC.
(18)
JOHNS,A.G. 1997 Timber production and biodiversiry conservation in tropical rain forests. Cambrige University Press,
Cambridge. (19)
JOHNS,
J.S., BARRETO,
P. and UHL,C. 1996 Logging damage during
planned and unplanned logging operations in the eastern
Amazon. For: Ecol. Manage. 89: 59-77. (20)
G.J.M. 2000. Logging in south
JONKERS,
W.B.J. and VAN LEERSUM,
Cameroon: current methods and opportunities for improvement. Internat. For Rev 2 (1): 11-16.
JUNK,
W.J. 1984 Ecology of the varzea, floodplain of the Amazonian whitewater rivers. In: SIOLI,H. (ed.) The Amazon:
limnology and landscape ecology of a mighty tropical river
and its basin. Monographiae Biologicae No. 36. Dr. W. Junk
Publishers, Dordrecht.
KILKKI,
R. 1992 Reduction of wood waste by small-scale log
production and conversion in tropical high forest. FAO,
Rome. (22)
LESLIE,A.J. 1977 Where contradictory theory and practice coexist. Unasylva 29: 2-40. (23)
MATTSON
MARN,H.M. and JONKERS,
W.B.J. 198 1 Logging damage
in tropical high forest. UNDPIFAO Working Paper No. 5.
Forestry Dept., Kuching, Malaysia. (24)
MASON,
D. 1996 Responses of Venezuelan understorey birds to
selective logging, enrichment strips, and vine cutting.
Biotropica 28: 296-309. (25)
MONTENEGRO,
F. 1997 Extraccidn forestall de bajo impact0 en La
Mayronga. Funcaci6n Forestal Juan Manuel DurinilITTO,
Quito, Ecuador. (26)
A. and VIANA,
NEPSTAD,
DC., BROWN,
I.F., LUZ,L., ALECHANDRE,
V. 1992 Biotic impoverishment of Amazonian forests by
rubber tappers, loggers and cattle ranchers. Advances in
Economic Botany 9: 1-14. (27)
NICHOLSON,
D.1. 1958 An analysis of logging damage in tropical
rain forest, North Borneo. Malay. For: 21:235-245.
PHILIPS,O.L., HALL,A.H., GENTRY,S.A., SAWYER,
S.A. and
VASQUEZ,
R. 1994 Dynamics and species richness of tropical rain forests. Proc. Nut. Acad. Sci. USA 91: 2805-2809.
PINARD,
M.A. and PUTZ,F.E. 1993 Reduced-impact logging as a
carbon-offset method. Cons. Biol. 7: 755-758.
PINARD,
M.A., PUTZ,EE., TAY,J. and SULLIVAN,
T. 1995 Creating
project in Malaysia. J. For: 93: 41-45. (28)
PINARD,
M.A. and PUTZ,F.E. 1996 Retaining forest biomass by
reducing logging damage. Biotropica 28: 278-295.
PRABHU,
R., COLFER,
C.J.P. and DUDLEY,
R.G. 1999 Guidelinesfor
developing, testing and selecting criteria and indicators for
sustainable forest management. CIFOR, Bogor, Indonesia.
(29)
REPETTO,
R. 1988 Overview. In: REPETTO,
R. and GILLIS,
M. Public
policies and the misuse offorest resources. Cambridge University Press, Cambridge. pp. 1-41, (30)
C. l998 Planes de manejo forestall y necesidades de
SABOGAL,
informaci6n para el manejo operacional, Memoria del
Simposio Internacional sobre Posibilidades de Manejo Forestal Sostenible en AmPrica Tropical. BOLFORICIFORI
IUFRO, Santa Cruz, Bolivia. pp. 135-147.
SIST, P., NOLAN,T., BERTAULT,
J-G. and DYKSTRA,
D. 1998
Harvesting intensity versus sustainability in Indonesia. For:
Ecol. Manage. 108: 25 1-260.
SIST, P., DYKSTRA,
D. and FIMBEL,R. 1998. Reduced-impact
logging guidelines for lowland and hill dipterocarp forests in
Indonesia. CIFOR Occasional Paper No. 15, Bogor, Indonesia. (31)
TERSTEEGE,
H. 1998. Biomass estimates for forests in Guyana and
their w e in carbon offsets. Consultancy report, Iwokrama
International Centre, Georgetown, Guyana.
53
TERSTEEGE,
H. and HAMMOND,
D.S. 1996 Forest management in
the Guianas: ecological and evolutionary constraintsbn timber
~roduction.BOS Newsletter 15: 62-69. (32)
UHL,C. and VIERA,I.C.G. 1989 Ecological impacts of selective
logging in the Brazilian Amazon: a case study from the
Paragominas region of the state of Parfi. Biotropica 21: 98106.
VANDER HOUT,P, l999 Reduced impact logging in the tropical
rain forest of Guyana. Tropenbos-Guyana Series 6.
Georgetown, Guyana. (33)
G.J.R. 1998 Reduced impact
VANDER HOUT,P. and VAN LEERSUM,
logging: a global panacea? comparison of two logging studies.
In Research in tropical rain forests: its challenges for the
future. Tropenbos Foundation, Wageningen. pp. 185-202. (34)
VIDAL,E.,
JOHNS,J.,GERWING,
J.J., BARRETO,
P,, and UHL,C. 1997
Vine management for reduced-impact logging in eastern
Amazonia. For: Ecol. Manage. 98: 105-114. (35)
WHITMORE,
T.C. & BROWN,N.D. 1996 Dipterocarp seedling
growth in rain forest canopy gaps during six and a half years.
Phil. Trans.Roy. Soc. London B 351: 1195-1203.
WINKLER,
N. 1997 Environmentally sound forest harvesting.
Testing the applicability of the FAO Model Code in the Amazon
in Brazil. Forest Harvesting Case-Study 8. FAO, Rome. (36)
WILLIAMS,
M. 1989 Americans and their forests. Cambridge
University Press, Cambridge.
54
COMMENT
On Tipper and de Jong on 'Quantification and regulation of
carbon offsets from forestry: comparison of alternate
methodologies with special reference to Chiapas, Mexico'
IAN G . ENTING
CSlRO Atmospheric Research, PM6 1, Aspendale, Victoria 3195, Australia
In a recent paper, Tipper and de Jong (1998) argue for the
need for carbon sequestration projects to be evaluated in
terms of cumulative storage, in units such as tC.years (tonnes
of carbon X years sequestered). The existence of such a
measure would be highly desirable for managing
sequestration projects. Unfortunately, the mathematical
analysis by Tipper and de Jong is based on a misrepresentation of the behaviour of the carbon cycle and this
invalidates the conceptual basis of their approach.
In order to demonstrate the problems, the Tipper and de
Jong argument is quoted verbatim (with numbering, and
[additional comment] added by the present author).
(b) The suggestion that a temporary carbon sequestration
can have the effect of an emission reduction or
equivalently can provide an offset that, when measured
in terms of integrated radiative forcing, cancels an
emission.
g 06
; o4
C
02
- - _ _- - - - - _ _ _ _ _
The total warming effect of a given emission is
determined by the cumulative presence of the greenhouse
gas in the atmosphere; in other words the product of
concentrations and time.
In the case of CO,, terrestrial and oceanic sinks take up
carbon previousely emitted over time. [The time history
of CO, from a unit emission is characterised by what
Tipper and de Jong call a 'depletion curve'. An estimate
of this curve, showing the depletion of a 1 t carbon
release, is shown in part (a) of Figure i .]
Assuming the dynamics of the carbon cycle remain
stable, most CO, emitted at the present will be absorbed
within 100 years and the cumulative radiative forcing
will be proportional to the area under the depletion
curve, expressed in tC.years.
Calculation of this area provides an estimate of the
cumulative carbon storage required to offset an emission
of 1 tC at the present time.
Tipper and de Jong repeat this argument in mathematical
terms in their Appendix 1.
The key aspects of this argument are:
(a) The use of integrated radiative forcing as a measure of
importance of the climatic importance of activities causing
global change. This is well-established scientifically. In
thefmofGIoba1 Warming Potentials (GWPs), integrated
radiative forcing has become the basis for defining an
equivalence between different gases for the purposes of
emission targets defined by the Kyoto Protocol.
FIGURE1 Atmospheric response to a l t emission of carbon.
This perturbation gradually dissipates, as shown by the
curve in part (a). The net rate of change of atmospheric
carbon is shown in part (b),the negative values indicating
a loss from the atmosphere. This net rate of change is the
gross flux from oceanic (and biotic) systems to the atmosphere (the dotted curve in part ( c ) )minus the gross from the
atmosphere to oceanic and biotic reservoirs (shown as the
dashed curve in part (c)).
On Tipper and de Jong
The remainder of this paper is devoted to showing that
point (b) is incorrect and that in terms of integrated radiative
forcing, no temporary carbon sequestration can offset a
carbon emission. Any measure for crediting temporary
sequestration as an offset for emissions will have to use a
criterion other than integrated radiative forcing. Any such
criteria would involve a move away from the approach used
in the Kyoto Protocol. International acceptance of such
modified criteria is unlikely to be acheived quickly, if ever.
In the Tipper and de Jong argument, points 1 and 2 are
essentially correct. However point 3, in the form used by
Tipper and de Jong, has the difficulty that the area under the
curve is essentially infinite. While most CO, is taken up
within 100 years, about 15% remains in the atmosphere for
millenia or more (e.g. Sundquist, 1985). For this reason
(among others), the most common way of comparing
greenhouse gas emissions, the GWP, introduces a time
horizon to limit the time period of interest. The equivalances
defined by the Kyoto Protocal are based on a 100-year time
horizon. Tipper and de Jong represent a depletion as a single
exponential, thus underestimating the atmospheric effect of
a fossil emission in a way that exacerbates the error due to
their later over-estimate of the biospheric effect.
The more serious problem with the Tipper and de Jong
argument is that point 4 is simply incorrect. It needs to be
replaced by:
55
For carbon emissions, the depletion curve is shown in
Figure 1 (a). This represents the net effect of the rate of
change shown in part (b), which is the derivative of the curve
in part (a). The origin of this behaviour is shown in part (c).
The rate of change is the gross flux from oceanic and biotic
systems minus the gross flux from the atmosphere to
oceanic and biotic systems. These fluxes are shown in part
(C)as perturbations from an equilibrium value Feq(with Fes
of order 100 GtCIyear). The flow of carbon from atmosphere
to oceans (shown as the dashed curve in part (c)) is initially
increased but subsequently decreases due to the decrease in
the atmospheric concentration. For the gross fluxes from the
atmosphere, the perturbation from the equilibrium flux is
propo&ional to- the perturbation in the atmospheric
concentration and so the gross flux decreases as the
concentration perturbation decreases. The carbon exchange
from the oceans to the atmosphere is shown by the dotted
curve, again as a perturbation from the equilibrium.
Immediately after the carbon emission, this flux is F but it
increases as the amount of carbon in the oceans increases. A
similar argument applies to any biospheric exchange due to
'CO,-fertilisation' .
4*. Calculation of this area provides an estimate of the
cumulative carbon reduction in the atmosphere required
to offset an emission of 1 tC at the present time.
The need for this distinction rests on two points:
Firstly, since the purpose of the calculation is to evaluate
the warming effects, the calculation needs to be in terms
of changes in the atmosphere.
Secondly, terrestrial storage is not equivalent to
atmospheric carbon reduction. For example, storage of 1
tC in the biosphere for 5 0 years does not lead to a 50 tC
cumulative carbon reduction in the atmosphere.
Implicitly equating the atmospheric effect to the
cumulative storage is to overestimate the effect of
sequestration.
However, when sequestration is evaluated according to
atmospheric changes (as it must be for purposes of
considering warming) the change due to a 1 tC (permanent
sequestration) is simply the negative of the change from a 1
tC emission. A detailed discussion of this point is given
below. This removes the possibility of using integrated
radiative forcing to equate a temporary sequestration to the
(permanent) effects of an emission.
The reason that 'terrestrial carbon storage' does not
equal 'atmospheric carbon reduction' is that in equilibrium,
if carbon from the atmosphere is sequestered in a forest, the
resulting lowering of atmospheric CO, leads to an imbalance
between the atmosphere and oceans s o that carbon from the
oceans tends to make up much of the loss. This is the 'mirror
image' of the way in which the CO, perturbation from an
emission is depleted.
FIGURE
2 Atmospheric response to a I t sequestration of
carbon. Again, the perturbation gradually dissipates, as
shown by the curve in part (a). The net rate of change of
atmospheric carbon is shown in part (b), the positive values
indicating a return of carbon to the atmosphere. This net
rate of change is the gross flux from oceanic (and biotie)
systems to the atmosphere (the dotted curve in part (c))
minus the gross from the atmosphere to oceanic and biotic
reservoirs (shown as the dashed curve in part (e)).
56
Enting
Figure 2 shows how sequestration leads to the reverse
process. A n intial lowering of atmospheric CO,
concentration leads to a reduction in the flux from
atmosphere to oceans (dashed curve in Figure 2 (c)). The
flux from oceans to atmosphere (dotted curve in Figure 2
(c)) is initially unchanged, but declines as carbon is lost from
the oceans to the atmosphere. The excess of gross flux to the
atmosphere over the gross flux from the atmosphere gives a
postive rate of change of atmospheric carbon as shown in
Figure 2 (b). This leads to a dissipation of the negative
perturbation in concentration.
Since the carbon dynamics is essentially linear at present,
the result for the reduced effect of carbon sequestration still
applies when considered as a small perturbation about large
fossil emissions.
Indeed, it is in some ways easier to understand the
difference between carbon storage and atmospheric carbon
reduction if it is considered against a background of larger
fossil emission. The arguments above, as illustrated by
Figures 1 and 2, show why the linearity (proportional
responses to proportional forcing) should apply equally to
positive and negative forcing. However, in practice, the
positive and negative changes are actually relative to an
overall positive fossil carbon input. Taking some simple
approximations:
If l00 tC of fossil carbon is emitted, then after l00 years,
30 tC will remain in the atmosphere.
If, through emission reductions, 90 tC is emitted (and the
other l 0 tC left in the ground) then after l 0 0 years 27 tC
will remain in the atmosphere.
If 100 tC of fossil carbon is emitted and 10 tC is
sequestered in a forest then after100 years, 27 tC will
remain in the atmosphere. Compared to the case of a 100
tC emission alone, the seqestration of I0 tC for 100 years
lowers the atmospheric carbon level by 3 tC at the end
of the 100 years, not by 10 tC.
To summarise: (a) the atmospheric response to a negatlve
perturbation is the negatlve of the response to a
corresponding positive perturbation, and (b) the atmospheric
effect of putting carbon in a forest is the same as leaving the
same amount of fossil carbon in the ground.
For the purposes of managing carbon sequestration
projects, crediting sequestration in tC. years has a number of
advantages, as discussed by Tipper and de Jong. Crediting
sequestration that may turn out to be temporary may also
have overall greenhouse benefits ~f the rate of climate
change in the early decades of the 21st century IS judged to
be a serious problem.
As noted by Tipper and de Jong, achieving these benefits
will require international agreement. As it stands, the Kyoto
Protocol does not provide any basis for crediting temporary
sequestration. The Tipper and de Jong approach is close to
the spirit of the GWPs used in the Kyoto Protocol. However,
it is so close that once the flaw in point 4 is corrected (and
finite time horizons introduced to remove the factors of
infinity) the Tipper and de Jong form of analysis produces
the 'Kyoto' outcome of 'no net credit for temporary
sequestration'.
A recent analysis by Dobes et al. (1998) has recognised
that defining a net credit for temporary carbon sequestration will require a significant shift of viewpoint away from
the GWP formalism. Recognising the equivalence between
sequestration and emission reduction, they sought to quantify
the benefits of temporary sequestration (or equivalently a
delay in fossil carbon emissions) in terms of reduced
radiative forcing using a differential formulation. This uses
the delay in integrated radiative forcing as the criterion for
defining 'equivalence'. This work must still be regarded as
a tentative exploration of one possibility. However, it is quite
certain that any proposal for quantified credits for temporary
carbon sequestration needs to be formulated in a way that
correctly represents the behaviour of the atmosphere.
REFERENCES
DOBES.L., ENTING,
J.G. and MITCHELL,
C.D. 1998 Accounting for
carbon sinks: the problem of time. In: DOBES,L. (ed.) Chapter
13 of Trading greenhouse enzissions: some Australian
perspectives, Chap. 13. Bureau of Transport Economics,
Canberra.
SUNDQUIST,
E.T. 1985 Geological perspectives on carbon dioxide
E.T. and BROECKER,
W.S.
and the carbon cycle. In: SUNDQUIST,
(eds.) The carbon cycle and atmospheric CO,: natural
variations Archean to present.. Geophysical Monograph No.
32, American Geophysical Union, Washington D.C., pp 5 - 59.
R. and DE JONG,B.H. 1998 Quantification and regulation
TIPPER,
of carbon offsets from forestry: comparison of alternate
methodologies with special reference to Chiapas, Mexico.
Commonw. For: Rev. 77 ( 3 ) : 219 - 228.
Reply to Enting
KTIPPER and B. DE JONG
Institute of Ecology and Resource Management,
University of Edinburgh, Darwin Building, Mayfield
Road, Edinburgh EH9 3JU, U.K.
E l Colegio de la Frontera Sur, Dept. Agroecologia,
Division de Sistemas de Production Alternativa,
Apartado Postal 63, San Cristobal de las Casas, C.P
29290, Chiapas, Mexico.
Enting correctly highlights an error in our calculation of the
amount of terrestrial carbon storage required to offset the
emission of a unit of CO, at the present time. While we
acknowledge the flaws in ihe methodology described in our
earlier paper, we maintain the view that the most appropriate
and practical measure of the overall climate change
mitigation impact of forestry projects is the cumulative
carbon storage, measured in tonne-years, As pointed out by
Enting, the error in our methodology consisted of the
omission of the feedback effects of a new carbon sink on the
Tipper and de Jong
magnitude of other biophysical sinks. These feedback
processes will tend to reduce the 'cooling' effect of an
addition to the growing stock of terrestrial carbon. However,
we suggest that once these effects have been corrected, the
cumulative carbon storage will still provide a more practical
approach than the alternative methods for quantifying the
effect of offset projects. The alternative methods are:
based on the annual uptake of carbon (tC/year),
based on the average change in stocks over the 'long
term' (tC).
The annual uptake (or emission) of carbon is the unit
used to report national contributions to global warming and
is the basis of targets for reducing climate change. For these
purposes, it is an entirely obvious and appropriate measure.
However, it is important to distinguish between national,
regional or indeed biome or ecosystem carbon budgets and
projects whose purpose is to offset or compensate for the
emission of a given amount of greenhouse gas at a given
time. The use of annual uptake-based accounting for carbon
for offset projects could have a number of negative
consequences:
-
-
-
it would be very difficult to match the uptake of a forestry
project with a series of emissions from one or several
businesses -forestry activities take up little or no carbon
in their first years, followed by a period of relatively
rapid uptake, and, depending on the management, there
may be subsequent releases that must be debited in some
way.
companies could become over-dependent on forestry
offsets, since it would not be possible to purchase oneoff offset to cover a particular emission (e.g. for an
amount above a target). It would be necessary to invest
in a stream of carbon uptake over a period of 30 to 50
years.
fast growing plantations are likely to be chosen in
preference to the restoration of natural forest ecosystems
-while such plantations may have rapid uptake for 20 to
40 years they may have lower long-term carbon storage.
The average long-term change in stocks of carbon has
been used as a measure of the impact of a number of forestry
offset projects. However, there are several difficulties with
its practical application:
i.
For many forestry systems it is difficult to predict what
the long-term average carbon storage will be, since
equilibrium may take centuries to achieve.
ii. There is no definite time horizon for contractual or legal
purposes.
iii. How should one compare a project with an low average
carbon storage of 40 tC ha-', achieved within 20 years
with a project giving an increased carbon storage of 200
tC ha-' but over 150 years?
In comparison with these methods, the measurement of
cumulative carbon storage within a designated time-frame
appears to offer a more practical approach. In essence, it
recognises the temporal nature of the climate change
Reply
57
problem and values the delay of emissions through the use
of terrestrial carbon storage until technological alternatives
to fossil fuel can be deployed.
The application of cumulative carbon storage accounting
can be readily illustrated by the following case: assuming we
accept a 100 year time-frame, an emission of 1 tC would be
compensated for by the immediate uptake of l tC and storage
for l00 years (giving l00 tC.years cumulative storage). Any
warming due to this source would thus be delayed for at
least 100 years. At the end of this period the carbon offset
provider is released from any legal obligation and the
responsibility for any subsequent emissions is passed to the
decision makers alive at that time. The project is, in effect
a transaction with the future generation, who receive no net
warming and an intact (hopefully viable) forestry system.
The main theoretical objection to this approach is the
adoption of an 'arhitrary' 100 year cut-off point. This
effectively means that all damage due to the emission within
the 100 year period is valued equally (discount rate of zero),
but any damage after the 100 year limit is not valued
(discount rate of infinity). It is possible to make a strong
argument against the use of an arbitrary time horizon.
However, in practical terms, some such figure is almost
certainly required to define the legal obligations of carbon
service providers. Furthermore, it seems reasonable to adopt
a time-frame of this order that is manageable in terms of
forestry planninq and in terms of the rate of technical
innovation for the replacement of fossil fuel dependent
technology. Finally, this approach is consistent with the
established IPCC convention to compare the Global
Warming, Potentials of different greenhouse gases.
While the authors acknowledge the imperfections in this
method, they suggest that it remains of practical use for the
purpose of evaluating the offset value of a particular forestry
project.
58
REVIEWS
JEFFERY, R. and SUNDAR,
N. (eds.) A new moral economy for
India S' forests? Discourses of community and participation.
Sage Publications, Thousand Oaks, California and New
Delhi. 1999.304 pp. US ISBN 0-7619-9354-1 hbk, 0-76199355-X pbk; India ISBN 81-7036-820-0 hbk, 81-7036-8219 pbk.
This volume is one output of an Edinburgh UniversitylIndian
Council for Forest Research and Education research project on
joint forest management (JFM) in India. It evolved out of a seminar
convened to bring together recent research in the country that
could help provide a fresh perspective on the very large and
important set of initiatives to change forest management that have
been attempted under JFM programmes.
An extended introductory essay by the editors provides a
valuable review and discussion of historical factors that underlie
theevolution of JFM, and why the decision that such aradical shift
in the way forests should be managed was needed. The other
principal thrust of this paper is to critically examine the concepts
of 'community' and 'participation', and the ways in which these
have been interpreted and deployed in debates about joint
management. This provides both an interesting overview of the
broader context within which JFM is located, and a useful
framework for the papers that follow.
The first group of papers looks at the concept of community in
more detail. In an original conceptual treatment of the subject,
Arun Agrawal shows how normative assessments of community
have changed over time, and how they have varied between
different theories of social and economic change. The concept of
the traditional community has as often been associated with
arguments that they represent values and forms of organisation that
need to be changed, and with resistance to such change, as they
have with propositions that the community provides a vehicle for
change. Agrawal also argues the need to distinguish between
community as shared understandings and community as a form of
social organisation, and the dangers of failure to do so. Other
papers in this group, by Guha and Sivaramkrishnan, examine the
history of community and forest management and control in
particular regional situations.
Most of the rest of the papers examine contemporary
developments based on participation and community in particular
forest situations. In an account of an eco-development exercise,
Baviskar shows how differently the concept of community
participation has been interpreted by park authorities and villagers.
In a study of a forestry programme in Orissa Savyasaachi argues
that villagers' concepts of the forest are not recognised at all.
Vasavada et al. examine the pros and cons, and consequences, of
different government departments each seeking to establish its
own committee within a community, in order to ensure local
participation in its programme. The three papers that follow all
examine aspects of the large JFM project in the Western Ghats
region of Karnataka. Saxena and Sarin draw upon their in-depth
assessment of the programme as a whole, highlighting concerns
about the focus on degraded forests, a lack of a sense of
'ownership' of local JFM activities by village committees, and the
need for institutional change at forest department and government
levels. Correa examines the processes that lead to women and other
disadvantaged groups becoming marginalised as a result of the
programme, and Locke also focuses on gender issues.
In a final paper Bhaskar Vira turns to issues that JFM raises for
foresters, identifying factors that influence their acceptance or
resistance to the changes in the ways they are expected to relate to
and work with villagers. He argues for more flexibility in allowing
situation-specific local level arrangements to emerge, to reflect the
differences in both villager and forester agendas and possibilities
from place to place, and presents an interesting typology of
conditions under which joint management could work.
A collection of papers such as this inevitably raise issues of
coverage, content and internal coherence. As the editors point out
there are no contributions reflecting research into the ecological
issues related to joint forest management. Some of the papers have
a micro rather than a macro focus. However, the book also
contains papers, starting with the thoughtful Introduction, that
provide a fuller and more thought-provoking view of the relevance
of the concept ofjoint management, and the forms it can take, than
are to be found in most of the literature on the subject. Given the
importance of JFM in India, and the extent to which other
countries look to it for guidance in developing joint forest
management approaches of their own, a critical examination of this
nature is to be warmly welcomed, and the book is recommended
to all those with an interest in participatory collective approaches
to forestry.
J.E.M. ARNOLD
LEONARD,
J-P. Contribution d la opologie des principaux
s y s t h e s forestiers. Essai de classification physionornique
des for2ts a partir des facteurs sociaux generateurs.
[Contribution to a typology of principal forest systems. An
attempt at a classification of forests based on social
determinants.] Ph.D. thesis of L'Universit6 de Bordeaux I11
- Michel d e Montagne, Geography Department. 1999. 472
pp. Mimeo.'
The fundamental point of this interesting thesis is easy to state:
wooded areas are no longer natural ecosystems but largely
anthropogenic. The author sets out to show that a forest system is
a complex, made up of the composition and form of the component
stands, land ownership, forest labour, markets, etc. abl subject to
the impact of socio-economic forces.
The writer, now in his 70s, is a former colonial forest officer
and a long time forestry adviser to the pulp and paper industry in
SW France who has travelled extensively and thought deeply
about forestry questions. He is therefore in a good position to
provide a view based on a very wide range of historic and current
examples drawn from all parts of the world. He sets out to describe
four types of forest system.
The first is the subsistenceforest,its primary purpose being to
supply food. Its characteristics are: tribal ownership and
management, harvesting by families, a wide diversity of flora and
fauna, wood as a product of secondary importance. It evolves to
I
We are especially glad to print this review because of the general
interest and importance of the subject which, however, is rarely
discussed. D.R. Johnston aired the topic in 1976 in Forestry in
a changing world, Forestry 44(1):30-44, see in particular pp.3840.
Reviews
produce a landscape with many clearings, often created with the
aid of fire and used either for shifting agriculture or extensive
cattle-grazing, leading to complex agroforestry or to small villages
on land quite separate from the wooded areas.
The second is the industrial forest where industry's requirements for a steady supply of standard wood are met by permanent
and specialised areas, or 'tree-farms', consisting of a few species
or even clones, cheap labour and substantial investment in land,
plantations, tending and infrastructure. The author gives two
examples: i. the forest serving France's iron industry in the 17th
and 18th centuries with large compartments of coppice of graduated
ages, ii. the charcoal- or pulp-orientated forest of central Brazil
today. In both, the industry dictates the appearance of the forest
and the whole system. Its future is uncertain unless it be towards
regular high forest furnishing sawlogs, as has been the case with
the maritime pine forest of the French Landes.
The third type is regular high forest, managed for the supply
of sawlogs, a model scarce outside Europe but now under trial in
various countries. In most cases, soil and stand have the same
owner, workers are highly specialised and the product is highpriced. To illustrate the point that socio-economic factors such as
the degree of urbanisation, national income per capita, etc.
strongly influence the fate of wooded areas through the influence
of demand for wood products, Leonard compares the rate of
conversion for coppice to high forest in France with the speed of
transformation to softwood plantations in Germany between 1815
and 1940.
These three models are consistent with a growing demand for
and rising price of wood. The fourth is more pessimistic and
slightly provocative. The future of wood use depends on price and
in particular that price at least keeps pace with harvesting costs.
Already one can find large areas no longer logged because of the
negative return on cutting: vide South Korea, Japan, Switzerland
and Germany. The author considers these 'uneconomic' wooded
areas under two headings: sanctuary forests such as sacred woods
in some countries or abandoned high forest left unmanaged in
others; and wildland forest as land no longer farmed becoming
colonised by trees. What will their future be? Greenery as a foil to
towns? Contrary to the author's view I think that such areas must
be managed, certainly in a different mood and manner from the
past, for recreation, protection of water supply, nature conservation
and so on.
Thus four models of forest systems, each with a coherent
anthropogenic forest observable through space and time, each
demonstrating the dominant power of social and economic forces
over nature. As Leonard says "The forest is the mirror of society".
He presents an interesting graph with population density on one
axis and GDP per capita on the other showing how different fields
are occupied by the different forest systems. Curves linking the
historical developments in certain countries (France, Germany,
Japan) illustrate the evolution through the four states.
The strength of the thesis, the clarity of the exposition, the rich
material drawn upon (there are 480 references in the bibliography)
and the detailed examples combine to make one feel the necessity
of reading it in the original language. The author opens a revealing
window on the destiny of forests and, of course, the forestry
profession. It naturally leads one to wish to search for the future
of society as the key to decidingpresent forest policy. Nevertheless
some awkward questions remain. Are four models really enough
to describe the global situation and its forest systems? What about
the boreal forests, or the contrasts between the North and South
Islands of New Zealand? Despite the'neat picture represented in
the populatiodGDP graph how do other forests fit into that
59
picture, are not contrasting systems found in a single country at the
same time? Is progressive transformation from one system to
another easy, or even possible, or does it require a sudden shock
and involve a high cost? The work is certainly stimulating: like old
soldiers it seems that old foresters never die, they simply sprout
new shoots and provide us with new theories.
MITCHELL,
A.L. and HOUSE,S. David Douglas - explorer
and botanist. Aurum Press, 2 5 Bedford Avenue, London
WClB 3AT. 1999. xiii + 241 pp. £19.99 I S B N 1-85410591-4.
This book offers a very readable account of the life and background
of David Douglas. While mostly describing his travels and the
plants he encountered, it also provides insights into society in the
early 19th Century and reminders of hardships endured by many.
The book falls into two main parts. The first twelve chapters
describe his life as a chronological narrative. The second part is
a review in one chapter and four appendices, summarising his
achievements. The first chapter sets the scene of an artisan family
providing a strong ethical grounding for a somewhat impatient
schoolboy, From such a beginning, Douglas developed a keen and
curious eye for everything around him and long-lasting desire to
study and add to his knowledge. (In late-20th century terms,
'Continuous Professional Development' seemed to come naturally
to him!) He started work at the age of l l as an apprentice gardener
at Scone Palace. Eight years later, he moved to Valleyfield in Fife
and then to the Botanic Gardens is Glasgow. His capability for
hard work and quickness of uptake led at the age of 24 to his
appointment as a plant collector by Horticultural Society (not then
elevated to its present 'Royal' status).
Accounts of David Douglas's three collecting expeditions
form the bulk of the book. His first, in 1823, was to the east coast
of America and lasted six months. The second was to the Pacific
north west via Cape Horn, and took nearly three and a half years.
The success of this expedition and the importance of his discoveries
and introductions are the foundation of his continuing fame. Then,
in 1829, after year in Britain, he set out on his longest expedition
to the California and some of the Pacific islands. In 1834, while
visiting Hawaii, his life was tragically brought to a violent end
when he was just 35 years old.
In describing his voyages, the authors have struck a nice
balance. By using extracts from Douglas's own journal, his
Douglas's personal characteristics shine out - courage and
endurance, energy, sharp observation and botanical skills. The
authors fill in much background, partly quoting other contempo&y
records and partly commenting based on their own researches.
The authors have selected-their quotations and pointed their
comments for a predominantly forestry audience. However, they
also make it abundantly clear that Douglas also collected numerous
herbaceous and woody shrub species; many of these thrive in
gardens not just in Britain but in temperate regions wherever
horticulture is practised. Two appendices contain comments on
the more notable trees and garden plants; a third appendix lists
plants discovered, introduced or named by David Douglas. These
constitute an important part of the British exotic flora.
The book can be warmly recommended.
J.R. ALDHOUS
60
MILES,A. Silva: the tree in Britain. Ebury Press and Felix
Dennis Books, Random House, 2 0 Vauxhall Bridge Road,
London S W l V 2SA. 1999. 400 pp. £30. ISBN 0-09186788-6.
The idea of a 'coffee table' book is to impress the viewer and Silva
certainly impresses. Whether Evelyn would be pleased to have the
present writers trade on his title is less clear, especially as it
concerns trees much more than forests. For though filled with facts
about trees, the ways in which man has used them, painted them
and generally admired them, the book is not a treatise or a polemic.
There are 11 chapters ranging from 'The evolution of forest and
woodland', through 'Wild fruit and orchards' to 'Conservation
and tree planting'. The material is thus wide ranging, for the most
part accurate though the story line is never quite clear. The book's
sponsor, Felix Dennis, tells us in his preface that he became
obsessed by trees and the book illustrates this fully. His principal
author has been helped by John White, arboriculturist, on tree
species selection, development of the British tree flora and
conservation, by Anne McIntrye on folklore and medicine.
Stephen Daniels on trees in art. The photographs are stunning and
these are the real joy of the book and its claim to coffee table status.
The print quality is high and pages where text is set against a
dark background to the point of illegibility are mercifully rare. The
index is sensible. There is some repetition, e . g . pp.259 -261 on
rowan, some curious inclusions, such as tamarisk, introduced for
the sake of its medicinal properties, an extraordinarily eclectic list
of trees proclaimed as alternative species for particular situations
(p.343), and some odd claims are made such as that (p.351)
concerning Leyland cypress 'Not a pleasing environmental
prospect, but almost certainly a commercially viable one', or that
on sycamore (p. 354) to the effect that 'grey squirrel has succeeded
in almost destroying this alien tree' (where does the writer live?).
Misprints are few and far between. Page 349 has '/10' where nigra
after Juglans should appear. But the real point of noticing this book
with its slightly maverick arrangement and content is not to carp
but to recommend it as the ideal present for anyone interested in
countryside and trees' contribution to the British scene.
A.J. GRAYSON
NEWMAN,M. F., BURGESS,P. F. and WHITMORE,
T. C.
Malesian Dipterocarps. Foresters' CD-ROM manual. Royal
Botanic Garden Edinburgh, Edinburgh. 1999. £20.
High quality systematic monographs are one of the bases for
understanding botanical biodiversity. Unfortunately, the
accessibility of monographs to an audience wider than professional
botanists is limited by their availability and the complex botanical
language in which they are often written. This CD-ROM on the
Dipterocarpaceae goes some way to making information on this
large, economically important family widely available. The CDROM is a compilation of seven previously published manuals for
foresters on the commercial big-tree Dipterocarpaceae species,
plus 20 additional species from Peninsular Malaysia. Of the 10
genera and 386 Dipterocarpaceae species in Malesia, 257 species
and 21 subspecies are covered by the CD-ROM. Much of the
taxonomic work is derived from Ashton's treatment of the family
for the Flora Malesiana project, and this has been used to produce
detailed descriptions of each of the species. These descriptions are
complemented by leaf and fruit drawings and by distribution maps.
For each of the species, information is given on ecology and
geographical range, whilst other parts of the text describe
silviculture, wood anatomy and utilisation. The use of Adobe
Acrobat Reader makes moving around the Manual simple
and convenient. In addition to making large amounts of information
on dipterocarps readily available, the CD-ROM also provides a
multi-access key for species identification and, where practical,
more traditional single access keys based largely on field
characters. Both types of key are very valuable features of the
current CD-ROM, and once the manipulation of the multi-access
key has been learnt it is relatively simple to use. This reasonably
priced CD-ROM should be a valuable resource for researchers
working in South East Asia.
THEROYALHORTICULTURAL
SOCIETYThe RHS Plant Finder
Reference Library - 2000 CD-ROM Professional version.
£80 (Standard version £29.99). ISBN 0-950048-4-8.
This disk contains the electronic version of The RHS Plant Finder
together with other gardening and horticultural databases.
There are two versions, standard and professional.
The standard edition contains:
The RHS Plant Finder 1999 -2000 (over 70,000 plants and
their suppliers in the UK)
The RHS Good Plant Guide (2000 entries with about 1000
colour photographs)
Dictionary of common names (70,000 entries)
Lexicon of Latin names
A UK Gardens database
Garden societies
Award of Garden Merit (a listing of over 6000 plants awarded
the RHS's AGM)
In addition the professional version includes
The PPP index (listing over 60,000 plants and their suppliers
in 16 European countries)
Plantenvinder (listing 45,000 plants and their suppliers in the
Netherlands)
Gardening by Mail 1999 (listing nurseries in the USA and
Canada who export plants)
Wholesale suppliers (900 UK wholesalers) - this is an addition
to this annual edition
NCCPG National Collection Holders
Cultivar Registers
Hillier Gardener's Guide to Trees and Shrubs
University of York/English Heritage Parks and Gardens
A full description of the contents of the CD can be found on
the website <http://www.plantfinder.co.uk> together with
instructions for obtaining a copy online as well as by phone, fax,
e-mail or post.
This is an impressive collection of databases that is a pleasure
to use. While many of us really prefer to thumb through a catalogue or turn the crisp pages of a book when searching for
information this disc will convert many to the advantages of
searching 'on screen'. Linkages between the main plant lists,
suppliers, lexicons etc. are well organised. For about half the 2,000
plants in the 'Good Plant Guide' excellent colour photographs are
included. All this is in the standard edition. This would satisfy
most foresters and plant enthusiasts who enjoy a bit of gardening
and need reminding of the wealth of plant material available, their
names and synonyms.
Reviews
The professional edition is for specialists who need to trace
plants in Europe and North America and contact wholesalers in the
UK. The onlysections that most readers of this review would miss
is the complete text of Hillier's 'Gardener's Guide to Trees and
Shrubs' (though this is not linked to the main plant list). There is
little point in purchasing the disc for the Internet Directory of
Botany. This can be accessed directly at the Botanical Museum,
Finnish Museum of Natural History at <http://www. helsinki.fi//
kmus/botmenu.html>. This was updated to the end of l998 with
about 4,000 botany related links. The site includes a form for
sending corrections, additions and comments. For this type of
information direct access to the Internet is more efficient than
having out-of-date data on CD.
It is hard to comment on content of this magnitude. Spot
checks have not failed to produce verifiable information - from
availability of an ornamental tree from a local nursery to a source
for an unusual apple variety. Of the 2000 or so varieties supposed
to occur in the UK some 900 are listed under Malus domestica in
the plant list. Some 75 varieties of Picea abies are shown, but these
include synonyms and varieties no longer available that were listed
in previous editions.
The section on Garden Societies covers organisations mainly
in the UK but also from 24 other countries. This is somewhat
uneven and includes some, but definitely not all, Wildlife Trusts
and conservation organisations not obviously related to
horticulture. It is intriguing to note that the future of the Brit~shIvy
Society is said to be 'somewhat uncertain as we went to press', but
good to know that 'Les Croqueurs de Pommes' of France has
2,500 members munching apples and protecting some of the 4,000
regional and local apple varieties in that country.
The list of Kew 'Authors and Genera' and the synonyms
indicated in the main list will be a boon to writers of papers
containing botanical references. The CD must have a wide appeal
and this reviewer has spent many hours browsing and discussing
his finds with other enthusiasts. The standard addition will be good
enough for most of us.
P.G. ADLARD
NSHUBEMUKI,
L., MWANSOKO,
H.J.M. and MUGASHA,
A.G.
Istilahi za Elimumisitu Kiingereza-Kiswahili [ F o r e s t p
Terminology English - Kiswahili]. Forest Research Support
in Tanzania FORST and Forest Research Institute of
Tanzania, TAFORI. 1999. Pbk 269 pp. ISBN 9987-58003-3.
This very welcome book addresses the difficulties sometimes
faced by foresters in the interpretation of English technical terms
and jargon. Its potential area of use is very wide, encompassing not
only Tanzania, but also much of Kenya, Uganda, Congo and
northern Malawi, Zambia and Mozambique. It is directed at the
non-English reader at the practical, field level but it will be
valuable, too, for those with a partial knowledge of English who
are not quite sure what an English term actually means.
The book is arranged as a straightforward dictionary, with
single word equivalents where appropriate, but with fuller
explanations where necessary. Although no references are given,
the reader might need on hand the s'tandard Swahili-Swahili
dictionary (Inst. of Kiswahili Research 19811 and also Johnson's
A Standard English-Swahili Dictionary Inst. (Interterritorial
language committee 1939). which contains many items about
forests, trees and people. Few plant names are included and I am
not aware, though I may have missed it, of an up-to-date
61
replacement for Part 1 of the Check-Lists of Forest Trees and
Shrubs, No 5 (Hora 1940) which contained many kiswahili and
local equivalents of Latin plant names although the book had
errors and was never reprinted. Williams' (1949) book on plants
of Zanzibar and Beentje's (1994) on Kenya trees, shrubs and lianas
also have some information on Kiswahili names, but none of these
books is readily available to the target readership of the book under
review.
This new forest terminology therefore advances both the cause
of forestry and the Kiswahili language, the former based on the
expertise of the first author who is the acting Director of the
Forestry Research Institute. The second author brings to the
dictionary the authority of the Institute of Kiswahili Research at
the University of Dar es Salaam. Anyone working with forests and
forestry in the swahili-speaking countries needs a copy of this
book, which is well printed, strongly bound and genuinely pocketsized. I compliment the authors and suggest they might consider
looking into the need for a companion volume on Latin-KiswahiliEnglish plant names for the region.
P.J. WOOD
BEENTJE,
H.J. 1994 Kenya trees, shrubs and lianas. Nairobi,
National Museums of Kenya. 722 pp.
HORA,F.B. 1940 Checklists of the forest trees and shrubs of the
British Empire, No 5 Tanganyika, Part I Botanical and
vernacular names. Oxford, Imperial Forestry Institute.
INSTITUTE
OF KISWAHILJ
RESEARCH
l98 1 Karnusi ya Kiswahili Sanifu
[A Standard Swahili-Swahili dictionary].
OUP, Dar es
Salaam. 325 pp.
INTER-TERRITORIAL
LANGUAGE COMMITTEE 1939 A standard EnglishSwahili dictionan. OUP, London, Geoffrey Cumberledge.
548 pp.
WILLIAMS,
R.O. 1949 The useful and ornamental plants of
Zanzibar and Pernba. Zanzibar, Zanzibar Protectorate
Secretariat. 497 pp.
62
MISCELLANEA
Deliberate introductions of species: research needs
IPCC
The above is the title of a report published in 1999
(BioScience 49(8):619-630) of a workshop on species
introductions held in Hawaii. The sub-heading reads
'Benefits can be reaped, but risks are high'. All of us have
seen and read of dramatic examples of deliberate
introductions that 'got away' and this workshop managed to
bring sense to an enormous field in which powerful interests
are at work. As noted in the conclusions 'Although many lay
people have not given species introductions much serious
thought, those with economic, political, or professional
interests in the issue hold widely varying views. At the
extremes, these views range from a handful of advocates of
no introduction, or of such rigorous pre-introduction proof
of [benignity] that all introductions are effectively
prohibited, to an equally small group that advocates a
freewheeling global eco-mix of species. Happily such
extremists are now much in the minority; most proponents
of purposeful introductions understand the risks (but believe
that technology can deal with them), and most conservation
biologists recognize the potential benefits to be derived
from carefully controlled introductions.' This last sentence
includes bold words.
The workshop explored the areas of research that it
considered needed attention. Among these were the
following: research to better evaluate risks and benefits, on
alternatives to introductions, on purposeful introductions,
and work to evaluate and mitigate impacts of introductions.
In the last field are noted: breadth of impact of biological
control agents; evaluation of impacts on ecosystem processes
and services; post-introduction range expansions; postintroduction time lags; spontaneous hybridization;
genetically modifed organisms. The report can be
recommended as a valuable survey by a group of 21
biologists drawn principally from the plant sciences,
including forestry, who have thought deeply about a subject
of increasing concern as the volume of trade and the means
of transport multiply.
The Intergovernmental Panel on Climate Change (IPCC)
publishes a report on the vexed question of carbon and land
use change later in 2000; vexed because of the difficulties of
measurement, both in terms of technique and of principle.
Thus much anxiety surrounds the ability to say about a large
remote area of forest a. what the carbon balance is at some
starting date, b. what it has become as a result of man's
actions as opposed to 'natural' change. Equally opinions
differ in ways that surprise economists about the accounting
for trade in wood products and therefore the location of
release of CO, by decay or fire. There is a strong lobby that
would penalise producers of wood products who export,
rather than those who demand the goods. Clearly, as anyone
who has dealt with trade statistics in studies of consumption
knows, there are difficulties in tracing where products end
up and thus release CO,, but that does not mean that
producers are to 'blame': shades of paper makers and
sawmillers being treated like drug dealers!
Tax incentive
At the Kyoto conference on climate change, the European
Union agreed to reduce its greenhouse gas emissions by 8%
of their 1990 levels by the 2008-12. This reduction was then
apportioned according to member countries' levels of
economic activity and as a result of the UK accepted a
bindihg contribution of 12.5%. The British government has
undertaken to better this requirement by setting its own
target of a 20% redution in CO, emissions by 2010. It has
also instituted a climate change levy payable on firms'
purchases of electricity, gas and other fuels. The Paper
Federation of Great Britain, whose members represent
major consumers of power and hence fossil fuel, concluded
an agreement with government at the end of 1999 whereby
they undertook to reduce the industry's energy corasumption
between 1990 and 2010 by over 40%. As a consequence
firms will be given 80% discounts on the climate change
levy otherwise charged.
Carbon accounting
Tax incentives versus regulations
Historic land use change perspective
Workers at the Carnegie Institute of Washington, Stanford,
California have studied what the earth's vegetation would be
like today if it had been untouched by human activities
(report to the Ecological Society of America, August 1999).
They estimate that the effects of land use change, principally
forest clearance and soil disturbance, have been to cause the
release of 180 gigatonnes (10 ') of carbon. This is equivalent
to about 75% of the mass of carbon released through burning
of fossil fuels to date.
Resources for the Future reports in Resources, Fall 1999 on
the outcome of studies of the effectiveness of quantity as
opposed to price based controls of carbon emissions. The
Kyoto Protocol sets out quantity targets for emissions, as
reflected in the comment above. Resources notes that many
environmentalists and policy makers find the quantity-based
approach appealing since it offers a guarantee on the
emissions requirement. But at what cost? The alternative is
to charge emitters so much per tonne released and the
question that arises is whether this method, which offers
Miscellanea
policy makers, monitoring authorities and industries the
ability to limit the potential economic costs, would be as
effective as a straight quantity control. Research by W. Pizer
indicates that the gains to society from adopting a price
based scheme would be five times greater than the best
quantity based arrangement. Further, he finds that the best
method of all would be a hybrid systyem under which policy
would fix the target level of emissions allowable but let
emitters buy additional emission rights at a fixed price.
The RFF conclusion of course raises the old problem of
the rich emitters 'buying their way out' of international
obligations, a familiar theme among cognoscenti of the
politics of the climate convention.
The World Forestry Center
We are grateful to Dr Dennis I? Dykstra for permission to
print this description of the Center. Ed.
The present-day World Forestry Center traces its roots to a
forestry exhibit hall and museum constructed for the 1905
Lewis and Clark Centennial Exposition in Portland, Oregon.
For more than half a century Portland's Forestry Building
claimed the title of the world's largest log 'cabin'. On the
evening of August 17, 1964, a fire of undetermined origin
consumed the entire log building and all of its contents.
Within a few weeks after the 1964 fire, the mayor of
Portland called a meeting of industry and government
forestry leaders to consider the creation of a new forestry
educational facility. What grew out of that meeting was the
Western Forestry Center, a non-profit institution funded
through memberships and charitable donations and
dedicated to educating the public about the importance of
good forest management and the value of forests both
environmentally and economically. The Center, which was
located near the Oregon Zoo on city-owned land in
Washington Park, opened its doors to the public in June
1971. The showpiece of the facility is an educational exhibit
hall and museum housing both permanent and travelling
exhibits, a 20 m tall 'talking tree' that explains the
fundamentals of tree biology in five languages (English,
French, German, Spanish, and Japanese), an educational
store selling books, videos, and gift items made from wood,
and a memorial hall in which historically important forestry
leaders from around the world have been commemorated.
A dozen or so years after its creation, the board of
directors of the Center began to realise that the original
educational mission, focusing as it did on the forests of the
Western United States, was becoming too limited in a
rapidly shrinking world. Elementary and secondary school
teachers and students were asking for information about
forests in other parts of the world, especially the tropics.
Tourists were beginning to visit the Center in large numbers
from Europe, South America, and Asia, and they were
asking about forests in their own countries. Thus, in 1986 the
board decided to expand the organisation's mission to
encompass all forests of the world. To emphasise this
63
change, the institution was renamed the World Forestry
Center and the board of directors was expanded to include
overseas members. Board members now come from a total
of 15 countries. The 2000 board of directors.meeting will be
held in Finland.
In 1989, a subsidiary named the World Forest Institute
was created. The Institute complements the public-education
mission of the Center by focusing on professional-level
education and research. It does this primarily through an
Institute Fellow program in which young forestry and forest
products professionals from all over the world reside at the
Institute for one year, working on research projects of
interest to their sponsors and at the same time learning about
forestry issues in North America's Pacific Northwest. Their
research is published by WFI, and revenues from the sale of
publications are used to support the Institute's activities.
Fellows also have the opportunity to earn university credit
for their research through Oregon State University 'S College
of Forestry, and can take university-level courses in forestry,
forest products, or natural resource disciplines at Oregon
State or in other subjects (such as business and economics)
at nearby Portland State University. Fellows are commonly
sponsored by government institutions or the forest industry
in their home countries. The sponsorship funding required to
support a WFI Fellow for one year is US$20,000.
64
FORTHCOMING INTERNATIONAL EVENTS
March 20-25. Launceston, Tasmania, Australia. Future
of eucalypts for wood products. IUFRO 5.06.03 Improving
the utilization of plantations of eucalypts. Contact Gary
Waugh, CSIRO Forestry and Forest Products, Private Bag
10, Clayton South, 3169 Victoria, Australia., tel +61-39545-2122, fax +61-3-9545-2133,
email [email protected]
Website www.ffp.csiro.au/conference/iufro/
March 21-25. Nancy, France. Criteria and indicators for
sustainable forest managment at the forest management unit
level. Conference arranged by IUFRO, ECOFOR and the
European Forest Institute. Contact Olivier Laroussinie, GIP
ECOFOR, 19 av. du Maine, 75732 Paris Cedex 15, tel+331-45-49-88-36, fax +33-1-45-49-88-39,
-
April 5
8. Stellenbosch, South Africa. Workshop,
symposium and reunion on the theme of sustainable forestry
organised by the South African Institute of Forestry in
conjunction with the Tropical Forest Resources Group, UK,
the Faculty of Forestry of Stellenbosch and the Council for
Scientific and Industrial Research. i. April 5 - 6. Workshop
organised by the Tropical Forest Resources Group, UK
entitled 'Sustainable forestry: new imperatives in South
Africa. ii. April 7. Symposium entitled 'Sustainable
forestry: managing the risks and challenges', iii. April 8.
Millennium reunion. Contact Wim du Plessis tel +27 (0)
21 808 4812, fax +27 (0)21 808 3603,
April 6-8. Munich, Germany. Information management in
forest enterprises. IUFRO 4.04.02. Contact Martin Moog.
Chair of Forest Economy Science, Ludwig Maximilian
University, Munich, Am Hochanger 13, D-85354 Freising.
Germany, tel +49-8161-7164-30, fax +49-8161-7146-3 1.
April 9-14. Noosa, Australia. Symposium on hybrid
breeding and genetics. Contact Heidi Dungey, Queensland
Forestry Research Institute, MS 483, Fraser Road, Gympie,
Queensland 4570, Australia, fax +61-7-5482-8755,
email [email protected]~
April 24-29. Cuba. Manejo sostenible de 10s recursos
forestales. IUFRO 1.07.09. Contact Pastor Amador,
Universidad de Pinar del Rio, Faculdad Forestal, Marti
No 270, Pinar del Rio 20100, Cuba,
email dptopfor @netupr.upr.edu.cu
Website: http://iufro.boku.ac.at/iufro/iufronet
May 3-6. Vienna, Austria. Third International Conference
of the European Society for Ecological Economics
(ESEE) 'Transitions towards a sustainable Europe: ecology
- economy - policy'. Contact Klaus Kubeczko, Dept. of
Social Ecology, Institute for Interdisciplinary Studies
of Austrian Universities, P.O. Box 232, Seidengasse 13,
A-1070 Vienna, Austria, tel +43-1-526-75-01-20, fax +431-523-58-43, email [email protected]
[email protected]
Website http:Nwww.univie.ac.at/iffsocec
May 8-12. Edmonton, Canada. The role of forests and
forestry in the global carbon budget. Sponsored by the
Canadian Forest Service in conjunction with the
International Boreas Forest Research Association. Contact
Carbon Conference coordinator 5320- 122 Street, Edmonton,
Alberta, Canada T6H 3S5, fax +l-780-435-7356, email
carbon @ nofc.forestry.ca
Website http://www.nofc.forestry.ca/carbon
May 15-26. Nairobi, Kenya. 5th meeting of the Conference
of the Parties to the Convention on Biological Diversity.
Contact CBD Secretariat, World Trade Center, 393 Jaques
St., Suite 300, Montreal, Canada H2Y 1N9, tel +l-514-2882220, fax +l-514-288-6588, email [email protected]
Website http://www.biodiv.org
May 24-30. Lima, Peru. 28th session of the International
Tropical Timber Council. Contact ITTO
Website http://www.transport.com/-leje/itto.html
April 21- 29. New Delhi, India. India's forest beyond
May 27-30. Houghton, Michigan, U S A . 19th
International meeting for specialists in air pollution effects
on forest ecosystems. IUFRO 7.04.00 Impacts of air
pollution and climate change in forests. Contact David
Karnosky, School of Forestry and Wood Products, Michigan
Technological University, 101 U.J. Noblet Forestry
Building, 1400 Townsend Drive, Houghton, Michigan
4993 1-1295, U.S.A., tel +l-906-487-2898, fax +l-906487-2897, email karnosky @mtu.edu
2000? Commonwealth Forestry Association seminar, in
collaboration wiht the Indian Council of Forestry Research
and Education and the Department of International
Development, UK, on India's forest policy - management
strategies and institutional and organisational arrangements
in the light of local, national and global challenges. Contact
Dr VKBahuguna, tel and fax +91-11-4360379.
June 4-9. Leiden, Netherlands. International symposium
on the biogeography of SE Asia 2000. Contact Rienk de
Jong, Nationaal Natuurhistorisch Museum, Department of
Entomology, PO Box 9517, 2300 RA, Leiden, The
Netherlands, tel+3 1-7 1-5 16-26-52, fax +3 1-71-513-33-44,
April 17-May 5. New York, U S A . 8th Session of the
Commission on Sustainable Development.
Forthcoming international events
65
June 22-23. Lake Tahoe, Nevada, U.S.A. Wood adhesives
2000. IUFRO 5.00.00 forest products. Contact John A.
Youngquist, Project Leader, Performance Designed
Composites, USDA Forest Service, Forest Products
Laboratory, One Gifford Pinchot Drive, Madison, Wisconsin
53705, tel +l-608-231-9398, fax +l-608-231-9582.
Website http://www.fpl.fs.fed.us/pdcomp/
and communities. IUFRO 1.O7.O5 Natural regeneration of
tropical rainforests, 1.05.08 Natural stand regeneration ,
3.05.00/6.01.00/8.01.00/ 8.07.00. Contact Dr Natalino
Silva, Brazilian Agricultural Research Corporation, CP 48,
CEP 66240, BelCm, Par& Brazil, tel +55-91-226 6622, fax
+55-91-226 9845, email [email protected],
natalino @ amazon.com.br
June 25-30. Prolognan-la-Vanoise, France. Multipurpose management of mountain forest: concepts, methods,
techniques. Contact Gerard Buttoud, 14 rue Girardet, F54042, Nancy, fax +33-3-83-37-06-45, email buttoud
@nancy-engref.inra.fror Martin Parry, 11 Bevington Road,
Oxford OX2 6NB, U.K., fax +44-1865-284-69 1,
October 8-13. Durban, South Africa. Forest genetics for
the next millennium. IUFRO 2.08.01 tropical species.
Contact Colin Dyer, IUFRO Conference Organiser, PO Box
11636, Dorpspruit 3206, South Africa, tel+27-33 1-425779,
fax +27-331-944842, email [email protected]
October 29-31. Asheville, U.S.A. Third International Oak
Conference. Contact Ron Lance, The North Carolina
Arboretum, 100 Frederick Law Olmsted Way, Asheville,
NC 28806-9315, U.S.A., tel +l-828-665-2492, fax +l-828665-2371, email [email protected]
June 26-30. Wageningen, The Netherlands. 2000 World
Conference on natural resource modelling. Contact
Joost Meulenbroek, Congress Office, Wageningen
University, Costerweg 60, 6701 BH Wageningen, The
November 16-21. Washington D.C., U.S.A. National
Netherlands, tel +3 1-317-482029, fax +3 1-317-485309,
Convention, Society of American Foresters. Contact Mike
Website http:l/www.cqs.washington.edu/-gordie/rma/html Murphy, SAF, 5400 Grosvenor Lane, Bethseda, MD.208 142 198, tel +l-301-897-8270, fax + l -301-897-3690, email
July 16-21. Mkrida, Mexico. The impact of global
[email protected]
environmental change on forests, and the impact of forests
on global environmental change. Contact Eric Diaz, INIFAP,
November 29-December 2. Bhopal, India. Workshop on
Km.24 Carretara Mkrida-Motul, Mococcha, Yucatan CP
national level criteria and indicators for sustainable forest
97445, Mexico, email [email protected] or
management of dry forests in South Asia. Contact Patrick
Judy Loo, Canadian Forest Service, Box 4000, Fredericton,
Durst, Regional Forestry Officer, FAO Regional Office for
NB E3B 5P7, Cananda email [email protected]
Asia and the Pacific (RAP), 39 Phra Atit Road, Bangkok
10200, Thailand, tel +66-2-28 1-7844 ext. 139, fax +66-2August 5-6. Kuala Lumpur, Malaysia. Improved forest
280-0445, email [email protected]
management and harvesting for tropical forests. FAO/
IUFRO satellite meeting of the IUFRO World Congresss;
November 28-December 2. Kathmandu, Nepal. Bio3.05.00 Forest operations in the tropics. Contact R. Heinrich,
technology applications for reforestation and biodiversity
FAO Forestry Department tel +39-06-570 54727, email
conservation. BIO-REFORIIUFRO-SPDC. Contact [email protected]
REFOR, Nepal Workshop, c10 Nepal Flora Implementation
Project Office, Dept.of Plant Resources, MFSC, HMG/N,
August 6-10. Newfoundland, Canada. Annual conference
G.O.P. Box 2270, Kathmandu, Nepal, tel +977-1-25 1159,
of the Canadian Institute of Forestry. Contact Len Moores,
fax +977- 1-251141, email [email protected]
fax + 1-709-637-2290,
December 1. Washington D.C., U.S.A. Annual meeting of
August 7-12. XXl IUFRO World Congress, Kuala
the International Society of Tropical Foresters. Contact Dr
Lumpur, Malaysia. XXl IUFRO World Congress
Warren T. Doolittle, ISTF, 5400 Grosvenor Lane, Bethesda.
Organising Committee, 52 109 Kepong, Malaysia, fax +60Maryland 20814, U.S.A., email [email protected]
3-636-7753, email [email protected]
Website http:frim.gov,my/iufro.htm1
December 10 13. Canberra, Australia. 5th Pacific rim
bio-based composites sympsium. Contact Dr P. Evans,
August 21
25. Kuhmo, Finland. Restoration and
Department of Forestry, Australian National University,
management of biodiversity, 4th international workshop on
Canberra ACT 0200, Australia, tel +61 2 6249 3628,
disturbance dynamics in boreal forests. Organised by the
fax +612 6249 0746, email [email protected]
International Association of Vegetation Science and the
Website http://online.anu.edu.au/Forestrylwood/bio/
Finnish Biodiversity Research programme. Contact
bio.html
Workshop on disturbance dynamics, Department of
Forest Ecology, P.O. Box 24, FIN-00014, Helsinki, Finland
Website http://honeybee.helsinki.fi/dist2000
April 18-25. Freemantle, Australia. 16th CommonSeptember 20-22. Belkm, Brazil. New approaches to the
wealth Forestry Conference. Contact Libby Jones,
management of neotropical primary rainforest by industries
email [email protected],uk
-
-
66
ASSOCIATION AFFAIRS
Chairman honoured
The Government of Canada has appointed Dr Jag Maini,
chairman of the CFA and Coordinator of the Intergovernmental Forum on Forests of the United Nations, an
Officer of the Order of Canada. The Order is Canada's
highest civil honour and we are delighted to record Dr
Maini's achievement.
Farewell lunch to the Duke of Buccleuch
A lunch to honour the Duke of Buccleuch, immediate
Past President of the Association, who held that position
for 20 years, was held in Edinburgh on November 29, 1999.
Formation of United Kingdom Branch
At the meeting foreshadowed in the December 1999 Review
it was decided to form a UK Branch of the Association. The
chairman of the steering committee charged with establishing
the Branch is Dr Caroline Howard, with Mr Peter Branney
as Secretary, Mr Norman Jones as Treasurer, Mr Roger
Bradley (representing Scotland), Mr David Parsons (Wales)
and Mr Peter Wood (link with Executive Committee).
Annual General Meeting
The Association's AGM will be held at 2.30 p.m. at the
Oxford Forestry Institute on 19 May 2000. It will be
followed by a formal dinner at Green College, Oxford in the
evening. A technical meeting will be held on the 20 May at
O.F.I. at which young foresters will be given a platform.
New corporate member
We welcome as a new corporate member LTS International,
the Edinburgh based consultancy and project management
firm. LTS has been active in international consultancy in
forestry and conservation since 1973, in Africa, S and SE
Asia, Central / South America, Eastern Europe and Russia.
Staff at LTS include current members Julian Gayfer (who
has recently joined the company from DFID Delhi as
Director of Development Consultancy) and Ben Voysey.
Former member Alastair Fraser has recently returned to an
Edinburgh based role in consultancy and development of
new opportunities.
OBITUARIES
Laurence Roche, M.A., M.F., Ph.D.'
1927 1999
h u r e n c e Rwhe was a forester of international standing
w b a e career spanned three continents and four decades. He
i
brought to forestry a breadth of vision and an insatiable
curiosity about the individual's and society's relationship
with the natural environment. His international outlook was
fostered by his early life in the merchant navy and a period
of extensive European travel. As he often recounted, it was
the result of a chance meeting during his travels to Wales that
he decided to pursue a career in forestry. In 1956 he
registered for a forestry degree at Trinity College, Dublin
where he became strongly involved in debating and
developed his renowned powers of persuasion which proved
so useful in his career.
The first eleven years of his forestry career were spent in
Canada. He completed a Master's degree in Forestry at the
University of British Columbia followed by 4 years as a
research forester with the British Columbian Forest Service
during which time he completed his Ph.D. He then moved
to Quebec as a research scientist with the Canadian Forest
Service. Here he felt a growing desire to broaden his
horizons and gain experience of forestry in developing
countries. This ambition was realised when he was appointed
Professor of Forestry at the University of Ibadan, Nigeria in
1972. He had a profound influence on the development of
the department, broadening the scope of its activities by
launching new degree programmes in wildlife and fisheries
management and embarking on a vigorous programme of
staff development. The success of his research in
agroforestry at Ibadan led to his close involvement in the
foundation of the International Centre for Research in
Agroforestry in Kenya. He served as Vice-chairman of the
Centre's Board of Trustees and its Programme Committee
until 1987 and maintained strong links with the centre right
up to his death.
The longest period of his professional career was as
Professor of Forestry at the University of Wales, Bangor,
from 1975 to 1992. Under his stewardship, forestry at
Bangor underwent profound and enduring changes. He
raised the department's international profile particularly
with respect to the developing world and tropical forestry.
The curriculum was broadened to include more elements of
tropical forestry and a new undergraduate degree in
agroforestry and new masters' programmes in
Environmental Forestry and Forest Industries Technology
'
Professor Roche, a long standing supporter of the Association
-his last direct contact with us was as recently as the AGM held
at Boughton House in May 1999 - died after a short illness at
his home in Co. Limerick, Ireland.His many friends spoke of his
personal kindness. An example comes from a post-graduate
student from Ethiopia, Dr Bekele Gessesse, who recalls Larry
Roche coming out of his way from a mission in Ethiopia to a
place called Shambat, Khartoum to collect documents by hand
and subsequently facilitate Dr Gessesse's admission to Bangor.
Our sympathies are extended to his widow, Felicity, and their
children. Ed.
Association affairs
were introduced. Research activity and funding, particularly on tropical forestry and agroforestry topics, grew
rapidly. The fact that this expansion was achieved in a
period of increasing financial stringency is testament to his
political skills in convincing the University of his vision.
Roche firmly believed that many of forestry's problems
could only be solved through a holistic approach to rural
land use. In Bangor he argued for and in the late 1980's
achieved his aim of amalgamating two separate land use
departments into a School of Agricultural & Forest Sciences.
This created the only UK university department covering
agriculture, agroforestry and forestry.
A third notable contribution to both the university and
the international forestry community was his role as founding
editor of the Journal of Forest Ecology and Management.
This has become one of the world's most prestigious
international forestry journals and its success owes much to
his skilful guidance in its early years.
Throughout his period at Bangor he was a tireless
international traveller both in his promotion of Bangor and
through his work for FAO, ICRAF, UNDP, the World Bank
and other aid agencies. Recognition of his deep interest in
international forestry and his diplomatic skills resulted in his
election as President of the International Union of Societies
of Foresters from 1979 - 1984. Following his retirement
from Bangor the remained active in international forestry
working as a consultant for FAO and British, Irish and
international development agencies.
His many achievements in forestry would never have
been realised without his outstanding personal qualities. His
many friends around the world will remember him for his
humanity, constant encouragement, his clarity of vision and
sound judgement, his integrity and above all his sense of
humour and fun. The forestry world is poorer with his
passing.
R.J. COOPER
67
Mediterranean Development Project. In 1959, Jack was
awarded the Master of Forestry degree at the University of
British Columbia and immediately went on to complete a
second masters degree at Oregon State University in 1960.
Following a two year period as Assistant Professor of
Silviculture at New York State University College of
Forestry at Syracuse, Jack spent two years as Professor of
Silviculture in the College of Forestry, University of Liberia.
In 1967 he returned to Syracuse to commence studies
towards his Doctoral degree which was awarded in 1970.
Jack joined the Faculty of Forestry, University of British
Columbia as an Associate Professor in 1968 and was
promoted to the rank of Professor in 1982. He taught
silviculture, forest history and policy, and international
forestry. His research centred on the reclamation of badly
disturbed sites - a field in which he developed an international reputation. Jack had a passion for forest history and
was an enthusiastic member of the Forest History Society.
Jack published numerous scientific and professional
papers throughout his varied career, however, perhaps his
most important accomplishments were his books 'Man and
the ~ e d i t e r r a n e a nforest: a history of resource depletion'
published by Academic Press in 1981 and 'Cyprus: a
chronicle of its people, lands and forests' published by
University of British Columbia Press in 1987.
Jack retired from the University of British Columbia in
1989 and returned to live in Rothbury, Northumberland.
During his retirement Jack continued to take an interest in
forestry: his passion and concern for forest management was
reflected in a stream of comment on and criticism of UK
forestry and he played an active role in local land use issues.
He is survived by his wife, Olive, three children and five
grandchildren.
Jack was a storehouse of aphorisms. A favourite was
that: 'A professor is someone who thinks otherwise'.
Everyone who knew Jack, knows that he more than lived up
to this ideal.
D. HALEYand G . F. WEETMAN
Jack Thirgood, BSc., M.F., Ph.D.
1924 1999
-
Jack Vincent Thirgood was born in Newcastle and lived in
Whitley Bay, Northumberland as a child. After serving in
the Royal Air Force he attended the University of Durham
and then the University College of North Wales at Bangor
where, in 1950, he was awarded his B S c in forestry and
botany. Upon graduation he joined the Forestry Commission
and was assigned to the Research Branch where he remained
for four years. During this period he worked on the
reforestation of coal mining tips and developed a life-long
interest in land reclamation and the reforestation of highly
degraded sites. From 1954 to 1956 he worked for the
Cyprus Forest Service as a research silviculturist and helped
to initiate aResearch Division. 1956-57 found Jack working
for the FAO as the Director of the National Forest Land
Research Centre in Iraq and later in Rome on the
68
lntemationul Forestry Review 2(1), 2000
TRANSLATIONS: SUMMARIES
FRENCH
La pratique de I'EFIR : introduction des techniques
d'exploitation forestiere B impact r6duit dans une
operation forestiere commerciale en Guyane
S. ARMSTRONG et C. J. INGLIS
Les auteurs analysent les implications commerciales de
I'introduction des techniques d'exploitation forestibre B impact
rCduit (EFIR) dam une vaste operation de rCcolte de bois. Les
rCsultats obtenus dans une zone exptrimentale de 800 ha en
Guyane indiquent que des mCthodes de rCcolte amCliorCes peuvent
limiter les dommages et gCnCrer des Cconomies, particulitrement
en utilisant des cartes de localisation des arbres avant la rCcolte.
L'application de saines pratiques de rCcolte exige une formation et
une supervision adkquates. Les principaux obstacles h
I'implantation de I'EFIR sont la disponibilitt et la stabilitt d'une
main-d'aeuvre qualifiCe, qui travaille dans des conditions difficiles.
L'adoption de nouvelles technologies et de systtmes d'information
et de "
nestion amCliorCs devraient aussi accroitre llefficacitC
opCrationnelle et limiter les dommages dans d'autres types
d'opCrations forestieres commerciales comme la construction de
routes.
Le r6Ie de l'exploitation forestiere B impact r6duit dans
la domestication de la forCt ombrophile n6otropicale
N. R. DE GRAAF
L'exploitation forestitre B impact rCduit n'est qu'un Cltment du
processus d'exploitation de la foret ombrophile nCotropicale.Pour
atteindre un rendement soutenu et une production Cconomiquement
viable, il faut aussi modifier la structure et la composition de celleci par un procCdC appelt domestication. Sa transformation en for&
jardinCe, une structure d'amknagement hautement dksirable,
pourrait s'effectuer B court terme en adaptant le Systtme sylvicole
CELOS. L'auteur analyse les rCsultats de recherches sur le CELOS
mendes depuis 1967, au Surinam, et prCsente un exemple
d'amenagement et de sylviculture pratiquCs par une compagnie
forestitre amazonienne certifiee FSC.
Avantages, obstacles et incertitudes lies B I'application
des techniques d'exploitation forestiere B impact r6duit
en zone tropicale.
D.S.HAMMOND, P. VAN DER HOUT, R. J. ZAGT, G.
MARSHALL, J. EVANS et D.S. CASSELLS
I1 a CtC d6montr6 que I'exploitation forestibre B impact rCduit
favorisait I'environnement en reduisant les dommages causCs aux
peuplements forestiers et au sol ; elle n'est toutefois qu'un des
&ltmentsd'une saine gestion forestibre. L'adoption de I'EFIR est
sufiont ik5e &desprbecupations relatives aux colits de production
et aux possibilitts de bCnCfices. Bien que plusieurs colits et
b6nCfices directs de I'EFIR aient &tCquantifits, bon nombre de
questions et d'obstacles potentiels restent Bexaminer. La formation,
les hausses salariales, le contrBle, le suivi et le bois perdu n'ont pas
toujours C t t pris en compte dans la comparaison des coilts de
I'EFIR et des systbmes d'exploitation forestitre conventionnelle.
Bien que les bCnCfices directs qu'offre I'EFIR en reduisant les
dCchets paraissent universels, ceux d'autres mCcanismes
gCnCrateurs de revenus, tels que la compensation de carbone, sont
moins Cvidents. Dans les rCgions privilCgiant les objectifs de
gestion B court terme, les contraintes institutionnelles et
Cconomiques continueront de restreindre les possibilitCs
d'implantation de I'EFIR.
L'exploitation forestisre dans le sud du Cameroun :
mithodes utilis6es et amiliorations possibles
W.B.J. JONKERS et G.J.R.VAN LEERSUM
Les optrations forestibres dans les for&ts ombrophiles du
Cameroun mCridional doivent etre adaptkes aux conditions
physiques, biotiques et socio-Cconomiques dominantes. Dans les
zones d'exploitation, le terrain est souvent abrupt et accident& La
foret renferme beaucoup d'arbres de trts grande taille, mais il
s'agit rarement d'essences marchandes. L'exploitation de la foret
devrait etre planifiie et rCalisCe en ttroite collaboration avec la
population pour qui cette ressource est importante. En dCpit de
sCrieux efforts pour mieux gCrer la foret, les opCrations forestitres
ont peu changC au cours des dernibres annCes. Comme on abat
moins d'un arbre par hectare, l'exploitation forestitre fait peu de
dommages. Environ 30 % du volume de bois abattu demeure dans
la forSt. Cenains ClCments de I'exploitation forestibre B impact
rCduit comme le treuillage ainsi qu'une formation et une
supervision amCliorCes, peuvent rtduire substantiellement les
dommages, les dtchets et les effets nCgatifs de l'exploitation sur la
population locale et la faune. Toutefois, d'autres ClCments tels que
la coupe de lianes et l'abattage directionnel sont moins adaptis aux
conditions dominantes.
Les le~onsd'une exploitation forestiere B impact rkduit
en terrain accident6 B Sabah, en Malaisie
M. A. PINARD, F. E. PUTZ et J. TAY
Entre 1992 et 1997, dans le cadre d'un projet pilote de
compensation de carbone, quelque 2 400 ha d'une vieille for&tde
diptCrocarp6es du sud-est du Sabah ont t t t exploit6s sulvant les
principes de l'exploitation forestibre B impact rCduit (EFIR). La
planification de la rtcolte, la coupe des lianes, l'abattage
directionnel et les restrictions relatives au debardage ont contribuC
B rCduire, dans les zones d'EFIR, la proportion de tiges d'origine
endommagees dans les peuplements de 50ii 28 %,et les dommages
au sol, de 13 B 9 % de la superficie totale, par rapport aux zones
d'exploitation conventionnelle. La taille et la densite des essences
marchandes des peuplements rCsiduels Ctaient sup6rieures dam les
Tranlations: summaries
zones d'EFIR, tandis que les facteurs de conservation de la
biodiversitk et de production durable de bois y Cvoluaient
positivement. Les terrains abrupts et I'imprtvisibilitt! des pCriodes
stches nuisaient au dCbardage terrestre. En pCriodes de pluie, de
grands volumes de bois Ctaient inaccessibles et la production Ctait
retardie. Dans cette rCgion, I'adoption d'une mCthode de debardage
aCrien permettrait d'augmenter la proportion de superficie rtcoltte
sans nuire B l'environnement.
L'exploitation forestikre a impact reduit en zone
tropicale : objectifs, principes et impacts
P. SIST
L'auteur dtcrit les objectifs et principes majeurs des techniques
d'EFIR en zone tropicale et passe en revue les rtsultats des
principales recherches concernant I'impact de I'exploitation
forestiere sur les peuplements residuels et la dynamique forestiere.
L'objectif premier des techniques d'EFIR est de rtduire
substantiellement la perturbation des sols et de la vtgttation
rtsiduelle par rapport B I'exploitation forestitre conventionnelle.
L'EFIR repose principalement sur une planification et un contr6le
serrCs de toutes les opkrations de rCcolte. L'importance des
dommages dus B l'exploitation forestitre dtpend non seulement
des techniques utilistes mais aussi de l'intensitt de l'exploitation.
En Afrique, ceux-ci sont gCnCralement faibles, l'intensitt de la
rCcolte dtpassant rarement 2 arbres par hectare. A l'autre extrCme,
en Asie du Sud-Est, l'intensitC atteint en moyenne 9 arbres par
hectare et peut endommager la moitiC des peuplements. A de telles
intensitts, les techniques d'EFIR ne peuvent rtduire les dommages
de f a ~ o nmarquCe.
Verification de I'applicabilite de I'exploitation forestikre
a impact reduit dans les for& de greenheart en Guyane
P. VAN DER HOUT
L'exploitation silective de la for& en Guyane diffkre peu de celle
qui se pratique ailleurs en zone tropicale. Les compagnies
forestikres rCcoltent un petit nombre d'essences et negligent le
reste du peuplement et ainsi, les rendements futurs. Les intensitts
d'exploitation sont plut8t faibles ( 5 m' ha-'), sauf dans les for&ts
de Chlorocardium rodiei (greenheart). Les rendements y sont
ClevCs parce que les tiges rCcoltables se prCsentent souvent en
bouquets. I1 en rCsulte de grandes trouCes dam le couvert. Le vaet-vient des dtbusqueuses dCtruit la rtgCnCration des essences
marchandes autour des souches et compacte le sol, ce qui
compromet davantage le rttablissement de la foret. Un systtme
d'exploitation forest&e B impact rtduit (EFIR) a CtC mis a" point
pour risoudre ces probltmes environnementaux. L'exploitation
forestikre conventionnelle (EFC) du greenheart a peu de chances
d'Ctre durable. L'EFIR a permis de rtduire de 65 % la superficie
endommagke lors du dtbardage, et de 40 %, la superficie moyenne
des trouCes d'abattage. L'abattage concentrk, en EFC, n'a pas
rCduit la perte de couvert par rapport B I'abattage plus disperst en
EFIR. Malgrt5 les avantages de I'abattage concentrt, en EFC,
I'application de I'EFIR n'a pas fait augmenter les co13ts
d'exploitation, en raison d'un rendement plus ClevC par hectare et
d'un dkbardage plus efficace.
69
SPANISH
RIL real: Introducci6n de tecnicas de tala de impacto
reducido en una operaci6n forestal comercial en Guyana
S. ARMSTRONG y C. J. INGLIS
Se evallian las consecuencias comerciales de introducir la Tala de
Impacto Reducido (RIL) en una operaci6n a gran escala de cosecha
de madera. Los hallazgos se basan en un ensayo de 800 hectareas
en una operaci6n de cosecha de madera en Guyana, e indican que
el datio puede reducirse y que se logran ahorros financieros
mediante la implementaci6n de pricticas mejoradas de cosecha,
particularmente a travts del uso, previo a la cosecha, de mapas de
ubicacidn de Arboles. El entrenamiento y la supervisidn adecuada
son esenciales para implementar una buena prictica de cosecha.
Las principales barreras para implementar el RIL son la
disponibilidad y la retencidn de gente con las habilidades
necesarias para trabajar bajo condiciones dificiles. La adopcidn de
nuevas tecnologias y sistemas mejorados de informaci6n y de
administraci6n deberian tambitn meiorar la eficiencia operational
y reducir el daiio en otras Areas de operaciones de la ingeniena
forestal comercial tales como la construcci6n de caminos.
La tala de impacto reducido como parte de la
domesticacidn del bosque lluvioso neotropical
N. R. DE GRAAF
La tala de impacto reducido es solamente una parte del proceso de
usar el bosque Iluvioso neotropical. Para lograr una producci6n
sostenible y econdmicamente viable, la estructura y composici6n
del bosque tiene que cambiar, un proceso que puede llamarse
domesticacidn. La selecci6n de la estructura del bosque es
altamente aconsejable para la administracidn y parece ser
alcanzable en el futuro cercano mediante la transformaci6n del
bosque a travCs de una adaptaci6n del sistema silvocultural
CELOS. Se discuten 10s resultados de investigaci6n de CELOS en
Surinam desde 1967, junto con un ejemplo de la administraci6n y
silvicultura de una compafiia maderera comercial y certificada por
FSC en la regidn Amaz6nica.
Los beneficios, cuellos de botella e incertidumbres en la
utilizacidn e instrumentaci6n de las tCcnicas de taIa de
impacto reducido en regiones pantropicales
D.S. HAMMOND, P. VAN DER HOUT, R. J. ZAGT, G.
MARSHALL, J. EVANS y D.S. CASSELLS
La tala de impacto reducido ha mostrado ser medio ambientalmente
beneficiosa a travCs de la reducci6n de 10s dafios al 10s Arboles y
suelos del bosque, pero es solamente un componente mAs de una
buena administracidn forestal. La instrumentaci6n del RIL depende
ampliamente de satisfacer las preocupaciones acerca del costo para
el productor y las incertidumbres acerca de 10s beneficios.
Mientras muchos de 10s costos y beneficios directos asociados con
RIL ha sido cuantificados, existen adn un ndmero de preguntas no
respondidas y cuellos de botella potenciales que necesitan
70
tenerse en cuenta. El costo de entrenamiento, alteraciones en las
demandas salariales, el monitoreo y verificaci6n de la produccidn
maderera no siempre han sido tenidos en cuenta en las
comparaciones de costo entre RIL y 10s sistemas convencionales
de tala. Mientras 10s beneficios directos de emplear RIL a travb
de la reduccidn de desperdicios puede ser considerados universales,
10s beneficios derivadados a travCs de otros planes de generacidn
de ingreso, tales como secuestro de carbono, no estan tan claros.
Las restricciones institucionales y econ6micas continuaran para
reducir la probabilidad de la implementacidn del RIL en aquellas
regiones donde estos promuevan objetivos administrativos a corto
plazo.
Operaciones madereras en Camerun del Sur: 10s
metodos actuales y las oportunidades para mejorar
W.B.J. JONKERS y G.J.R.VAN LEERSUM
Se requiere adaptar las operaciones de ingenieria forestal en el
bosque lluvioso de CamenIn sur a las condiciones fisicas, bidticas
y socio-econdmicas prevalentes. El terreno utilizado para las
operaciones madereras es frecuentemente inclinado y dificil. El
bosque contiene muchos Brboles bastante grandes, aunque pocos
de ellos son especies propicias de comericalizar. El bosque es un
recurso importante para la poblacidn, y la ingenieria forestal
deberia planificarse y ejecutarse en estrecha consulta y cooperacidn
10s pobladores. Mientras se han hecho muchos esfuerzos para
mejorar el manejo del bosque, las operaciones de tala han
cambiado poco en recientes afios. Se cosecha menos de un Brbol
por hectarea y por 10 tanto el dafio por tala es limitado. Cerca de
un 30% del volumen de madera cortado se deja en el bosque.
Algunos elementos de tala de impacto reducido, como tales como
el uso del cabrestante y una mejor instruccidn y supervisi6n,
pueden reducir sustancialemte 10s dafios de la tala, desperdicio y
efectos negativos sobre la gente local y la fauna silvestre, pero al
mismo tiempo, otros elementos, tales como corte de liana y el
derribo direccional, son menos apropiados bajo las condiciones
prevalentes.
Las lecciones aprendidas de la implementaci6n de tala de
impacto reducido en terreno montaiioso en Sabah,
Malasia
M. A. PINARD, F. E. PUTZ y J. TAY
Entre 1992 y 1997, cerca de 2400 ha de bosque dipterocarpo
maduro en el sudoeste de Sabah fue cosechado de acuerdo con 10s
lineaminentos generales de las tCcnicas de impacto reducido (RIL)
como parte de un proyecto piloto de sequestro de carbono. La
planeach de la cosecha, el corte de enredaderas, el derribo
direccional, y las restricciones para deslizar. contribuyeron a una
reducci6n en el daiio a 10s Brboles desde 50% al 28% de 10s tallos
originales; el daiio a1 suelo se redujo desde el 13% al9% del area
total en RIL en comparaci6n con las areas de tala convencional.
Los hileras remanentes en areas de RIL tuvieron una mayor
estmctura vertical y mejor almacenamiento de especies madereras
comerciales aue en areas taladas convencionalmente, con
ganancias positivas para la conservaci6n de la biodiversidad y
sostenibilidad de la producci6n de madera. El terreno empinado y
la carencia de periodos secos pronosticables fueron las bareras
para ef sistema de deslizamiento sobre el terreno, resultando en
grandes voldmenes de madera en sitios inaccesibles, y en demoras
en la producci6n provocadas por el clima de hBmedo. La
introducci6n de un sistema aCreo de cosecha in esta regi6n podria
permitir una mayor proporci6n de las areas ha ser cosechadas de
un modo medio ambientalmente aceptable.
Impacto reducido de la tala en 10s tr6picos. objetivos,
principios e impactos
P. SIST
Se describen 10s principales objetivos y principios de las ttcnicas
de RIL en 10s trdpicos y se presenta una revisidn de 10s resultados
mBs importantes de la investigacidn llevada cab0 sobre el impacto
de la tala en las hileras remanentes y la dinamica del bosque. El
objetivo principal de las tCcnicas de RIL es reducir sustancialmente
10s dafios a1 suelo y a la vegetaci6n residual en comparacidn la tala
convencional. El RIL se basa principalmente en una estrecha
planeacidn y control de todas las operaciones de cosecha. La
cantidad de daiio de la tala depende de no solamente las ttcnicas
utilizadas sino tambiCn la intensidad de la tala. En Africa, el dafio
de la tala es generalmente ligero debido a que la intensidad de
cosecha raramente excede 2 &boles por ha. La imagen extrema se
encuentra en Asia sur oriental donde la intensidad de tala es en
promedio de 9 Brboles por ha; esto ademis de el dafio que puede
afectar a la mitad de las hileras de bosque. Con tales intensidades
de tala, las ttcnicas de RIL no son eficientes en reducir
significativamente 10s dafios.
Probando la aplicaci6n de la tala de impacto reducido en
bosques de coraz6n verde en Guyana
P. VAN DER HOUT
La tala selectiva en Guyana difiere un poco de la tala en otra parte
en 10s tr6picos. Las compafiias madereras cosechan un ndmero
pequefio de especies sin poner mucho cuidado en el dafio a las
hileras restantes y de ende a las producciones futuras. Las
intensidades de tala tienden a ser bajas (+5m3 ha-l), pero en 10s
bosques de Chlorocardiurn rodiei (coraz6n verde), se obtienen
altas producciones porque 10s tallos aptos para cosecha tienden a
darse en grupos. Por consiguiente, se forman grandes brechas en
el dosel de bosque. Las operaciones de acarreo cercade 10s tocones
destruyen la regeneracidn comercial y compactan el suelo
affectando aun mas la recuperacidn de bosque. Se disefid un
sistema de tala de impacto reducido (RIL) para cubrir estos
problemas ambientales. El sistema de tala convencional (CL) de
coraz6n verde no es probablemente sostenible. El RIL fue exitoso
en reducir 10s dafios de acarreo en un 65% (area de terreno
disturbado) y para reducir el tamafio promedio de brechas de
derribamiento en un 40%. La pCrdida total de cubierta de dosel no
se redujo debido a la tala en grupo en CL, contratrio a la tala mBs
esparcida en RIL. La implementacidn de RIL no condujo a un
aumento del costo de tala a pesar de que CL tiene ventaja por la
tala en grupo, debido a que se obtuvo una producci6n mBs alta por
hectarea y el acarrero fue mBs eficiente.
ABERDEEN
University of opportunity
FORESTRY EDUCATION OPPORTUNITIES
The University of Aberdeen founded in 1495 has been awarding degrees in Forestry
rince 1914
BSc (Forestry)
Firmly founded on a scientific base, this degree emphasises resource and business
management aspects relevant to both temperate and tropical forests. The honours
degree normally requires four years of study but students with diplomas or equivalent
qualifications are usually admitted direct into the second year.
Diploma/MSc
in Forest
Management,
4groforestry
Dr
Arboriculture
The diploma courses (9-months) and MSc courses (12-months) are designed primarily
for practising foresters from the tropics and subtropics. Programmes are flexible and
can be designed to meet individual needs, including the involvement of courses from
other land-use departments. Courses include Forest policy and organisation; Foresl
planning; Tropical forestry; Tree Improvement; Forest mensuration and inventory;
Rural developmentforestry and Agro-forestiy; Silvopastoralism; Silviculture; Forest
protection; Harvesting and Arboriculture.
Research
Degrees
These are available at both Master and Doctoral level: a wide-ranging research
programme is available encompassing both temperate and tropical forestry issues
Aberdeen and the North-East of Scotland contain a concentration of research establishments in Land Use
and Environmental matters unique in Europe: in addition to the departments within the University there is
the Macaulay Land Use Research Institute, the Institute of Terrestrial Ecology, the Rowett Research
Institute and the Marine Research Laboratory; combining the interests of all these is the Aberdeen Centre
for Land Use. The University is also the base of CEMP, the international Centre for Environmental
Management and Planning
Writefor further details to
Dr W Huw Parry
Department of Forestry
University of Aberdeen
58 1 King Street
Aberdeen AB24 5UA
United Kingdom
Telephone 01224-272677 Fax 01224 272685
Telex 73458 UNIABN 9
e-mail [email protected]
UNIVERSITY OF ABERDEEN

reviews - University of Oxford

Transcription

Similar documents

“LUCIAN GRIGORESCU” SCHOOL MEDGIDIA – CONSTANTA

Asia Pacific Forestry Week 2016 Partnership for

Los combustibles fósiles liberan bióxido de carbono al quemarse e

SUSTAINABLE PACIFIC GROVE APRIL MEETING

Year 2013

Parque Nacional Natural Cueva de Los Guacharos

Wood you believe? - Georgia Farm Bureau