Amaro A., Reed D., Soares P. (editors) Modelling Forest Systems

Подождите немного. Документ загружается.

29Amaro Forests - Chap 25 25/7/03 1:54 pm Page 306

30Amaro Forests - Chap 26 1/8/03 11:54 am Page 307

26 Regulating the Yield of Goods and

Services from Forests: Developing

Tools to Support Management

Decisions and Policy

Development for Multiple

Objective Forest Management

P.R. van Gardingen

Abstract

Growth and yield models have been developed for many tropical countries with the intention

of supporting management decisions relating to the yield of timber from forests. These tools

have, however, largely failed to deliver signiﬁcant improvements in management practice.

Forest managers and policy makers increasingly have to consider the supply of a much wider

range of goods and services, including non-timber forest products and a range of environmen-

tal services. It is obvious that existing approaches for yield regulation, which have failed to

deliver management improvements for the regulation of timber, will be even less suitable for

application in systems of multiple objective forest management supporting the livelihoods of a

wide range of stakeholders. New tools and approaches are required to support decision mak-

ing for management and policy development.

The multiple objective forest management (MOFORM) cluster of research

(www.moform.org) aims to support the development of new knowledge and tools for the reg-

ulation of the yield of goods and services from tropical forests. This will involve the modiﬁca-

tion and extension of existing growth and yield prediction tools (models) to develop a yield

regulation toolbox. This will include tools for the prediction, allocation, monitoring and report-

ing of yield from forests linked with appropriate economic and ﬁnancial tools and instruments

supporting multiple objective forest management.

This chapter reports on the implementation of yield regulation pilot studies in Indonesia

and Guyana utilizing two growth and yield models MYRLIN and SYMFOR to support forest man-

agement decisions and the development of policy. The lessons learned from these studies are

discussed in terms of how to design tools and approaches that better meet the needs of forest

managers and policy makers for multiple objective forest management.

Centre for the Study of Environmental Change and Sustainability, The University of Edinburgh, UK

Correspondence to: P.vangardingen@ed.ac.uk

30Amaro Forests - Chap 26 1/8/03 11:54 am Page 308

308 P. van Gardingen

Introduction

Growth and yield models have been developed to support the management of tropi-

cal forests. The available models range in complexity from relatively simple size

class-based models to much more complex spatially dependent individual-based

models. Such models have been described and reviewed elsewhere (Vanclay, 1995;

Porte and Bartelink, 2002).

The availability of sophisticated growth and yield tools supporting the regula-

tion of yield of timber fr

om tropical forests seems to have done little to improve for-

est management, the livelihoods generated from forests, or conservation of tropical

forest resources. For this reason, forest managers and policy makers in tropical

countries often question the relevance of such tools. The current study aims to

describe processes by which growth and yield models may be applied more effec-

tively to support the generation of livelihoods from tropical forests and demonstrate

beneﬁts of such approaches to forest managers and policy makers.

This chapter is based on the results of two separate approaches. The ﬁrst

involved consultation with stakeholders in a range of countries to determine the

need for new r

esearch supporting the application of growth and yield management

tools in tropical countries. The second involved the application of existing growth

and yield tools for yield regulation planning for pilot studies in Indonesia and

Guyana

Demand for New Tools and Approaches

A process of consultation to deﬁne the needs for new tools and approaches support-

ing the regulations of yield of goods and services from tropical forests commenced

in 2001. A multi-stakeholder workshop supported by the Forestry Research

Programme (FRP) of the UK Department for International Development (DFID) was

held in Edinburgh, Scotland, in June 2001 to assess and document demand for new

research and dissemination activities.

The workshop identiﬁed a series of priority constraints. A common theme was

the need to make existing knowledge mor

e readily available to those responsible for

making policy and management decisions. This, combined with the lack of effective

ﬁnancial and economic instruments, was considered to be the main reason for the

failure of existing growth and yield tools to improve management practice in the

ﬁeld. The participants judged that there was little need to invest in further technical

development of existing growth and yield tools.

It was recognized that major challenges remain in many tropical countries

because of the lack of incentives for sustainable management of for

ests and the

prevalence of illegal logging activities. These factors combine to reduce the impact

of application of traditional approaches to yield regulation in the tropics for reasons

including:

• alternative land uses are potentially more proﬁtable;

• there are high opportunity costs associated with the adoption of improved (sus-

tainable) for

est management practices; and

• some options for sustainable forest management may maintain or increase the

mar

ginalization of signiﬁcant stakeholder groups, leading to increased conﬂict

and the potential for illegal harvesting or destruction of forest resources.

The ﬁndings of the June 2001 workshop were used to develop a concept for a

multidisciplinary r

esearch programme to support multiple objective forest manage-

30Amaro Forests - Chap 26 1/8/03 11:54 am Page 309

309 Regulating the Yield of Goods and Services

ment (MOFORM, www.moform.org). Three main topics of new research were sug-

gested (van Gardingen, 2002).

State of knowledge review on yield regulation

This review should consider methods for the prediction, allocation, control, monitor-

ing and r

eporting of the yield of goods and services from forests. The review should

document the needs for new approaches for yield regulation resulting from the

involvement of a wider range of stakeholders in forest management and the produc-

tion of a wider range of goods and services from forests. The review should also con-

sider how experience from temperate forests and developed countries can be adapted

and transferred to meet needs for tropical forest systems and developing countries.

Yield regulation toolbox

The need to adapt existing yield regulation tools to meet the needs of policy and

decision makers has been clearly identiﬁed. These include methods for the pr

edic-

tion, allocation, control, monitoring and reporting of the yield of goods and services

from forests and their linkage to other approaches used to support forest manage-

ment decisions, including ﬁnancial and economic instruments.

Financial and economic instruments

There have been a limited number of studies that integrate methods of yield regula-

tion with ﬁnancial analysis (McLeish and Susanty

, 2000; Ashton et al., 2001; Huth

and Ditzer, 2001; McLeish and van Gardingen, 2001). These studies considered the

return on investment for the production of timber using simple ﬁnancial cost–bene-

ﬁt analysis. There is a need to extend this work to consider a wider range of goods

and services as well as the economic beneﬁts of good forest management. The con-

sultation suggested a need for research to produce guidelines to which economic

and ﬁnancial tools and instruments are most appropriate under speciﬁc circum-

stances. This may require the development of methods that permit rational debate

between stakeholders who only use monetarized methods of valuation with those

who also consider social and cultural values. Suitable approaches that combine mul-

tiple criteria decision analysis and participatory methods have been developed for

use with other land-use systems and watershed catchment management (Proctor,

2001, 2002).

Designing Yield Regulation Pilot Studies

A series of yield regulation pilot studies have been implemented in Indonesia and

Guyana during 2001 and 2002. The locations for the pilot studies were chosen to

illustrate that each has characteristics that require site-speciﬁc approaches deter-

mined by factors including:

• the size and management objectives of the forest enterprise;

• availability of data describing the forest enterprise, including:

� static inventory,

30Amaro Forests - Chap 26 1/8/03 11:54 am Page 310

310 P. van Gardingen

� growth estimates from repeated growth measurements (permanent sample plots),

� spatial data (GIS) describing management compartments,

� ﬁnancial data;

• availability of suitably trained staff and local technical advisors.

The pilot studies were implemented in Indonesia and Guyana, representing con-

trasting for

est types and management regimes. Natural forests in Indonesia are domi-

nated by commercial dipterocarp species characterized by relatively high growth rates

and resulting potential productivity. In contrast, the majority of forests in Guyana are

on poor soils with low growth rates and commercial productivity. Both countries have

implemented systems of large industrial forest concessions for many years and have

recently started to encourage alternatives, including community forest management.

Data availability varied between locations. In each country, there was one loca-

tion wher

e a large industrial concession held information describing management

compartments, static inventory and permanent sample plots. These data permitted

the application of a range of growth and yield tools. One Indonesian site also had

detailed ﬁnancial information integrated as a model of the concession. In Guyana,

one community forest enterprise had limited information available, but in Indonesia

no suitable location could be found.

In both countries it was necessary to implement an extensive programme of

training for local partners to support the pilot studies. This training involved the

principles of yield r

egulation, application of tools and presentation of results to

other local stakeholders. Local researchers were trained to apply yield regulation

tools and then asked to design the studies for Indonesia (Ridwan and Redhahari,

2001) and Guyana (Khan and Singh, 2001).

Four types of pilot studies are described here. These are:

1. Analysis of policy options for industrial forest management in Indonesia.

2. Linkage of growth and yield models with ﬁnancial tools in Indonesia.

3. Yield regulation for large-scale industrial forest management in Indonesia and

Guyana.

4. Yield regulation for community forest management in Guyana.

The pilot studies were implemented to demonstrate how existing yield regula-

tion tools could support policy and management decisions and to identify how

these tools and their application could be further impr

oved. Two contrasting yield

regulation tools that could be applied in Indonesia and Guyana were available from

earlier research projects. These are,

MYRLIN (methods of yield regulation with limited

information, www.myrlin.org) and

SYMFOR (silvilculture and yield management for

tropical forests, www.symfor.org). The

MYRLIN toolbox was designed to produce a

simple growth and yield model to be applied at concession level, based on limited

data, speciﬁcally a single static inventory measurement. In contrast, application of

the

SYMFOR framework requires long-term data from permanent sample plots to cali-

brate an individual-based ecological model and is typically applied to describe

results of management interventions at plot or compartment level.

The

MYRLIN toolbox

The

MYRLIN toolbox is based around a simple diameter-class projection model

(Wright and Alder, 2000). The current version of the software and documentation is

available electronically (Alder et al., 2002).

MYRLIN is implemented as three simple

spreadsheet tools designed to assist the process of yield regulation in natural tropi-

30Amaro Forests - Chap 26 1/8/03 11:54 am Page 311

311 Regulating the Yield of Goods and Services

cal forest. These are: (i) the production of stand tables; (ii) estimation of growth and

mortality rates; and (iii) stand projection and harvesting model. These tools are

directed at the needs of organizations and enterprises that have only a limited

research base of information on forest dynamics for their locality.

MYRLIN is designed

to run with data describing a series of forest compartments (or management units)

describing part or all of a forest enterprise (typically a concession).

The

SYMFOR framework

The

SYMFOR framework was initially developed for mixed dipterocarp forests in

Indonesia (Phillips et al., 2002a) and includes spatially explicit, individual-based

models of ecological processes and forest management practice. New ecological

models have been developed for Guyana (Phillips et al., 2002b) and work is ongoing

with versions for forests in Brazil, Ecuador and Bolivia. The development of an eco-

logical model for the

SYMFOR framework requires data obtained from repeated mea-

surements of permanent sample plots. The model simulates the growth, recruitment

and mortality of each individual tree in a plot, typically 1 ha in size.

The model simulates a number of stochastic processes and thus should be run

epeatedly for each of several replicate plots in one forest type. A typical

SYMFOR run

may involve 20 replicate simulations for each of up to 20 plots. Such an analysis will

normally take at least several hours, but this will vary depending on the complexity

of the task, the number and replication of plots and the length of the simulation, as

well as the speed of the computer. Data output by the

SYMFOR framework then needs

to be subjected to further statistical analysis before being made available for use by

forest managers and policy makers.

Pilot Study Results

Study 1: policy analysis in Indonesia

The

SYMFOR framework was applied in conjunction with statistical analysis of forest

management data in order to evaluate the standard silvicultural prescriptions for

the dipterocarp forests in Indonesia. The model was programmed to simulate the

yield regulation system applied through the Indonesian selective logging and

replanting system (TPTI), which utilizes a diameter limit of either 50 or 60 cm for all

commercial species with a cutting cycle of 35 years (McLeish and Susanty, 2000).

The study showed the yield from the TPTI system was not sustainable and com-

pared it with a suggested alternative system of a maximum eight stems per hectare

(Sist et al., 1998; van Gardingen et al., 1998). This research concluded that a system

based on a maximum yield speciﬁed as either a maximum of eight stems per hectare

or a felled volume of 50 m

/ha was sustainable when combined with a 35–45 year

cutting cycle. These results were summarized in a short brieﬁng paper to make the

results more readily available to policy makers (Purnama et al., 2001).

Study 2: linkage of growth and yield models with ﬁnancial analysis

A number of studies have described the linkage of the

SYMFOR framework with a

concession-scale model of ﬁnancial performance (Fadilah, 1997) for the PT Inhutani

I concession in Berau, East Kalimantan, Indonesia. A general guide for the linkage of

30Amaro Forests - Chap 26 1/8/03 11:54 am Page 312

312 P. van Gardingen

the models (McLeish and van Gardingen, 2001) showed that a relatively simple

approach is adequate, which only requires estimates of the predicted yield per

hectare and number of stems from

SYMFOR and estimates of the costs of harvesting

and silvicultural treatments for the ﬁnancial model. This approach was applied in

an extensive study of yield regulation options for forests in the concession that com-

pared the TPTI system based on a 50 cm diameter limit with alternatives of eight

stems per hectare and maximum yields of either 50 m

/ha or 60 m

/ha (van

Gardingen et al., 2003). This study repeated the results of the previous policy study

(McLeish and Susanty, 2000; Purnama et al., 2001) showing that the sustained yields

could only be obtained when regulated using either maximum volume or number of

stems. The ﬁnancial analysis suggested that the best management option was likely

to be yield regulated to 50 m

/ha combined with a 35-year cutting cycle or 60 m

/ha

and a 45-year cutting cycle. Both of these management regimes gave an internal rate

of return that was predicted to exceed 10% over three cutting cycles.

Study 3: yield regulation for large-scale industrial forest management

The studies described above all involved the application of

SYMFOR to the analysis of

forest management scenarios for industrial forest management in Indonesia, repre-

sented by a limited number of sample plots from a single management unit and for-

est type.

MYRLIN was applied to a study of yield regulation at concession scale for the

PT Intracawood concession in Indonesia and the Barama Company Ltd concession

in Guyana. In both studies, data were derived from a GIS (geographic information

system) linked to stratiﬁed inventory data. These data were combined to produce a

description of the compartments to be managed. In each study, users calibrated the

MYRLIN growth and yield model using the spreadsheets provided in the toolbox

before running management scenarios to evaluate the sustainable yield. The users

found the toolbox easy to apply but noted that the management options were too

simplistic to allow the simulation of standard harvesting practice in either country.

The studies both recommended that the model required more ﬂexibility in manage-

ment scenarios and that the presentation of results required some indication of

uncertainty.

The application of

MYRLIN in Guyana was done in parallel to an application of

SYMFOR calibrated using sample plots from the same area. The two models produced

very similar results at low harvest intensities.

MYRLIN predicted lower yields than

SYMFOR when harvesting intensity was increased. This was attributed to differences

in the growth models since

SYMFOR models competition and increases the growth

rate of trees in the residual stand following harvesting, whereas

MYRLIN has constant

growth rates within a species group in all conditions.

Study 4: yield regulation for community forestry management

MYRLIN was applied to community forest management with the ITUNI small loggers

association (ISLA) in Guyana. This group had very limited information about their

forest managed under a woodcutting lease for an area of 33,300 ha. The ISLA are

required to produce a management plan for this area in order to secure long-term

access to the forest resources. Researchers from the Guyana Forestry Commission

assisted the ISLA to apply

MYRLIN to calculate estimates of sustainable yield for the

concession area. Data describing the forest were limited to basic vegetation maps.

These were improved through ﬁeld surveys, interpolation from soils maps, and a

30Amaro Forests - Chap 26 1/8/03 11:54 am Page 313

313 Regulating the Yield of Goods and Services

participatory mapping exercise. The resulting vegetation map was combined with

inventory data from an adjacent area of forest to provide data for the compartments

to be managed. The data were then used to calculate expected yield using

MYRLIN.

The results from this study predicted a sustained annual yield of around 14,000

for the concession based on a cutting cycle of 20 years and an annual coupe of

940 ha. This represented the ﬁrst estimate of likely production and will form a major

component of the management plan for the area. The study was implemented as the

ﬁrst step in an iterative process to improve yield estimates. The users of the results

identiﬁed a number of issues that need to be improved, including the collection of

inventory data from the concession and the implementation of more realistic har-

vesting options in

MYRLIN. It was noted by all those involved that the participation

of forest managers in the study ensured that the results were more relevant to man-

agement decisions and that the managers were more likely to act on the outcomes of

the study. One result of this process was the decision of the ILSA to implement their

own inventory for use in the next phase of planning.

Discussion

The pilot studies in Indonesia and Guyana have demonstrated how existing growth

and yield tools can be used to support management and policy making and sug-

gested ways to improve the effectiveness of this process. The studies were imple-

mented as an example of action research where, if possible, potential users of the

results were involved in the planning and implementation of the research. This

process has generated a number of lessons for future developments to build tools to

support decision making and policy development for tropical forests. These are

summarized here.

Adapt existing growth and yield models.

All of the studies suggested that the structure and design of the existing growth and

yield models wer

e adequate for the applications supporting management and pol-

icy development. The users did not identify any requirement for further develop-

ment of the biological and ecological content of the models they had used. Their

comments suggested that it was far more important for the models to be adapted to

improve the management options available and the method than that results were

presented to users.

The comments about management options mainly related to the available

methods of yield r

egulation that could be simulated. In

MYRLIN this is limited to har-

vesting a proportion of the standing volume above a user-speciﬁed diameter limit. It

was noted that this system could not simulate the current logging code of practice in

Guyana, nor could it simulate the suggested modiﬁcation to TPTI in Indonesia

based on a maximum number of harvested stems or volume.

SYMFOR was recognized

as having more ﬂexible harvesting options, but users sometimes found these difﬁ-

cult to implement. Users also expressed a need to extend the analysis of both models

to represent a wider range of goods and services provided by forests, initially to

non-timber forest products and then to include environmental services.

Users assessed the method of presentation of results from simulations as being

an extr

emely important part of the design of the tools. The graphical representation

of results by

MYRLIN was considered to be very helpful, but all users suggested that

this needed to be combined with some indication of reliability or uncertainty and

30Amaro Forests - Chap 26 1/8/03 11:54 am Page 314

314 P. van Gardingen

standard deﬁnition of terminology (in particular ‘annual allowable cut’). Users

found

SYMFOR more difﬁcult to use, as it required subsequent statistical analysis of

results before they could be used for management or policy decisions. It was sug-

gested that there is a need for more guidance to be provided on how to apply mod-

els to support management and policy (as opposed to users’ manuals) and on how

best to present results from simulation modelling.

In both countries it was possible to compare the two models. The simulated

yield estimates wer

e similar, but

MYRLIN tended to produce lower estimates when

competition decreased after heavy disturbance. This was considered a favourable

outcome, as

MYRLIN was designed to work with limited information and would pro-

duce conservative estimates of yield. It was noted that this did provide an incentive

for managers to collect more extensive data that could be used in more complex

growth and yield models. Of the models compared,

MYRLIN was well suited to oper-

ational concession-scale management, while

SYMFOR was best suited to more strate-

gic policy analysis. This emphasized that no one tool should be expected to meet the

needs of all potential users and that the choice of appropriate tools would depend

on the nature of the application and availability of data. The users expressed the

need for assistance in choosing appropriate tools for their applications.

Availability of data to apply growth and yield tools

The availability of data to develop or calibrate models and run simulations was

identiﬁed as an important limiting factor in all of the pilot studies. The choice of

locations for the studies was determined by the pr

esence of existing inventory or

permanent sample plots. The ISLA study demonstrated the importance of having at

least adequate inventory data for the area to be managed, and this was reinforced by

the inability to include a community forest management group for Indonesia.

Suitable data may not be available in many developing countries, especially with

the increasing trend for small-scale community involvement in forest management.

The ISLA study demonstrated how such groups could become involved in yield

scheduling planning using existing sources of information and encourage further

data collection.

Integration with other forest management tools

All of the studies demonstrated the beneﬁts of integrating growth and yield models

with other standar

d forest management tools used for planning purposes. The mod-

els need to interact with results from the static forest or compartment inventory and

this is often done via a GIS for the site. More sophisticated forestry operations may

also have a management information system (MIS) that contains information about

productivity and operating expenses.

The ﬁnancial analysis carried out in Indonesia relied on the availability of infor-

mation for the concession studied. The guidelines and user interface developed for

this study (McLeish and van Gar

dingen, 2001) demonstrated that integration of

tools requires the deﬁnition of protocols and standards for transfer of information

between systems. In the study reported here, the link was implemented through

manual data entry, but more sophisticated approaches may involve real-time inte-

gration of software. One example is the integration of a growth and yield model

with a GIS system for a forest concession as part of the Iwokrama project in Guyana

(www.iwokrama.org).

30Amaro Forests - Chap 26 1/8/03 11:54 am Page 315

315 Regulating the Yield of Goods and Services

Financial and economic instruments

It is essential to link growth and yield models with ﬁnancial and economics analysis

in or

der to be able to inﬂuence management and policy decisions. In nearly all situa-

tions, such decisions will be made in an environment comparing the management of

a production system with either an alternative production system or, more fre-

quently, alternative resource or land-use options. The process of decision making

should normally involve the consideration of at least ﬁnancial, economic, ecological

and social factors. The Indonesian pilot study demonstrated that it was possible to

design a forest management system that produced a reasonable return on initial

investment. This alone was, however, insufﬁcient to encourage changes in manage-

ment practice, as the study showed that the adoption of improved forest manage-

ment practice was associated with high initial opportunity costs and returned proﬁts

that were much less than alternative (albeit often illegal) management activities.

This indicates that there is a need for new research to guide which economic

and ﬁnancial tools and instr

uments are most appropriate under speciﬁc circum-

stances. One suggestion has been to explore the development of methods that allow

rational debate between stakeholders who only use monetarized methods of valua-

tion and those who also consider social and cultural values. Theses approaches

could then be applied to the aim of reaching a consensus on decisions that produce

the greatest net social beneﬁt. Research is also needed to consider compensation

mechanisms that have the potential to transfer value captured by downstream users

and consumers of goods and services, to beneﬁt those forest managers who adopt

good land and forest management practices upstream. This must include the devel-

opment of equitable mechanisms for the capture and distribution of beneﬁts (value)

resulting from good forest management.

Involve end-users in the studies

The pilot studies demonstrated the very real beneﬁts of involving end-users of

esearch results in the design and implementation of the studies. Their involvement

ensured that the research was more focused on the outcomes required to support

management and policy decisions. It also meant that feedback on the performance

of the tools reﬂected their needs rather than the academic interests of the

researchers. It is notable that none of the end-users requested improvements in the

biological content of the models used in the studies. Their emphasis was on making

the results more relevant to management and policy, through the provision of realis-

tic management options and effective (simple) presentation of results.

Present results to wider range of stakeholders

Forest management in most developing countries now involves a wide range of

stakeholder gr

oups. Management decisions and policy development should now

normally involve dialogue between these groups in order to reduce potential con-

ﬂict and develop a shared vision for the forest resources.

Results from the pilot

studies in Indonesia and Guyana were presented in multi-stakeholder fora that

included forest managers, policy makers, national and local government, civil soci-

ety groups, educational institutions and the media. The presentations made at each

forum stimulated debate between these groups, each of which had an interest in the

forest resource but had previously not engaged in debate about its management.

Accepting that it may not be possible to get all groups to agree to this vision.