Forest Sector Indicators - World Bankdocuments.worldbank.org › curated › en ›...

Forest Sector IndicatorsAn Approach for Central America

byMichael LinddalEnvironment Department, The World Bank

In collaboration withManuel Winograd, Andrew Farrow, and Marta AguilarCIAT-World Bank-UNEP Project, CIAT

With an Annex on water indicatorsConceptual Framework for the Development and Use of Water Indicators

prepared byManuel Winograd, Marta Aguilar, Andrew Farrow (CIAT), and Lisa Segnestam (World Bank)

November 2000

UNEP

THEWORLDBANK

40479

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Cover: Forest cover map for Central America. Dark areas are deciduous forests and lighterare coniferous forests. Source: CIAT-WB-UNEP 1999.

The judgments expressed in this report do not necessarily reflect the views of the World Bank, its Board of Directors, the govern-ments they represent, CIAT, or UNEP.

The boundaries, colors, denominations and any other information shown on all maps in this document do not imply, on the part ofthe World Bank Group, CIAT, or UNEP, any judgment on the legal status of any territory, or any endorsement or acceptance ofsuch boundaries.

The International Bank for Reconstructionand Development/THE WORLD BANK1818 H Street, N.W.Washington, D.C. 20433, U.S.A.

Manufactured in the United States of AmericaFirst Printing November 2000

iii

Contents

iii

Acknowledgments v

Abbreviations and Acronyms vii

Chapter 1 Introduction 1Objective 1The Rural Sustainability Indicators Project in Central America 1Outline 2

Chapter 2 Forest Indicators — Conceptual Issues 5Selecting Forest Indicators 5Demand and Supply of Indicators 6Candidate Indicators for the Forest Sector 8

Chapter 3 Regional Forest Indicators 11A Regional Core Set of Indicators 15

Chapter 4 The Forest Sector in Honduras 17Key Information on the Forest Sector in Honduras 17Developing Geo-referenced Forest Indicators for Honduras 18

Chapter 5 Possible Developments of Advanced Forest Sector Indicators 21A Wood Utilization Balance 21Carbon Accounts 22Asset Values of Forests 23

Chapter 6 Concluding Remarks 25

Annex A Maps of Forest Area and Related Issues in Honduras 27Annex B Conceptual Framework for the Development and Use of Water Indicators 31

Notes 37

References 39

iv Forest Sector Indicators — An Approach for Central America

Box

1 Forest sector characteristics and indicators 5

Figures

1 Demand and supply side of environmental indicators 72 The land expectation-, land-, and stumpage value of a forest stand 233 A project based approach 334 The Pressure-State-Impact-Response approach 34

Maps

1 Forest area in 1994 (including protected areas), Honduras 272 Forest area in 1985, Honduras 283 Forest area that was already lost in 1985, Honduras 284 Forest accessibility, Honduras 295 Forest and floods, Honduras 296 Forest fires, Honduras 30

Tables

1 Types of forest indicators 92 Some candidates of forest indicators 103 Forest area in Central America 114 Different sources of relative forest cover 125 Rate of deforestation — FAO data 126 Rate of deforestation — CCAD data 127 Matrix for changes in forest areas 138 Production of fuelwood and industrial roundwood 149 Trade in wood based forest products (1993) 15

10 Suggested set of feasible regional core indicators 1511 Forests areas in Honduras (1996) 1712 Removals of industrial roundwood in Honduras (1993–97) 1813 Sawnwood production and consumption in Honduras 2214 Final consumption of forest products in Honduras 2215 Country level natural capital estimates in Central America 2416 Examples of water indicators for a project-based approach 3317 Examples of water indicators for Pressure-State-Impact-Response approach 35

v

Acknowledgments

v

T his paper, Forest Sector Indicators: An Ap-proach for Central America, is a result of acollaborative project on rural sustainability

indicators for Central America between the Inter-national Centre for Tropical Agriculture, the WorldBank and the United Nations EnvironmentProgramme.

Thanks are due to Norberto Fernández, UNEP,and Lisa Segnestam and John Dixon, the WorldBank, for useful comments and suggestions on a

previous version of this paper. The capable assis-tance of Jim Cantrell with desktop publishing isgratefully acknowledged.

Neither this paper nor the project on ruralsustainability indicators would have been possiblewithout generous support from the Government ofDenmark, which is greatly appreciated, as is thesupport from the Governments of Norway andSweden.

vii

Abbreviations and Acronyms

vii

CCAB-AP Central American Council on Forests and Protected AreasCCAD Central American Commission on Environment and DevelopmentCIAT International Center for Tropical AgricultureCOHDEFOR The national forest agency in HondurasCUM Cubic metersFAO United Nations Food and Agriculture OrganizationFMU Forest Management UnitITTO International Tropical Timber Organizationn.a. Not availableNAI Net annual incrementNGO Non-Governmental OrganizationSFM Sustainable Forest ManagementUNEP United Nations Environmental Programme

1

CHAPTER 1

Introduction

T he exploitation of natural resources, andchanges in the services provided bythe rural environment, in Central America

and other developing regions have importantimpacts on the livelihood of rural communities.Uncontrolled development of forest resourcesmay result in growing poverty and loss of welfaredue to environmental degradation and depletionof social values and opportunities. If, on the otherhand, the possible trade-offs with environmentalservices and rural livelihood are taken intoconsideration when natural resources are ex-ploited, it may be possible to mitigate adverseimpacts on the environment and rural poverty.

Environmental indicators are a set of tools thatcan enable decisionmakers to link policy decisionsand regulation of natural resource exploitationwith possible adverse impacts and capability tomonitor those impacts. Access to this relativelyunbiased information can also become a crucialpart of consensus building and conflict resolution.

This paper presents an outline and discussionfor developing national indicators for the forestsector in Central America, focusing on Hondurasas a country case study. It is intended to presentan example of how indicators in a sector can bedeveloped on available data. Rather than acomprehensive analysis of the forest sector inCentral America based on empirical data, thepaper is primarily aimed at demonstrating anapproach for a selected sector as an input to alarger project on rural sustainability indicators inthe region.

Objective

Countries commonly take a sectoral approach tonatural resources, such as water, forests, coasts,and agriculture. One reason is that this is how thepublic administration is structured, and mostsector-oriented policies offer few incentives tolook at the cross-linkages between sectors, ecosys-tems, natural resources, and human welfare. Asector-based approach thus speaks to a specificpublic institutional setting. It may be an efficientway to reach the level of policy-making, but posesa risk of focusing on partial policy issues. Thechallenge therefore, is to make the best of thesector-based approach while ensuring thatindicators can capture inter-sectoral linkagesreaching beyond the sectors and their self-interest.

The objective of this forest sector example is todemonstrate how policy-relevant issues can beaddressed at the regional and national levelthrough the use of available forest sector informa-tion derived from existing data sources.1 Thesector example does not attempt to reach the locallevel where the key sector issues are related tomanagement. Moreover, it is not an attempt todevelop indicators where information is notavailable or relatively easy to retrieve.2

The Rural Sustainability Indicators Projectin Central America

This paper represents one component of a collabo-rative effort between the International Center forTropical Agriculture (CIAT), the United Nations

2 Forest Sector Indicators — An Approach for Central America

Environment Programme, and the EnvironmentDepartment of the World Bank.3 The project—rural sustainability indicators project in CentralAmerica—aims at developing and testing aframework of environmental indicators for ruraldevelopment in Central America. More specifi-cally, the objectives of the project are to:� Develop, test, and refine environmental, land

quality, and other related indicators andinformation tools in a user-friendly geographicinformation system interface, for theintegration of rural sustainabilityconsiderations into policy-making andplanning

� Improve environmental management at theregional, national, and local levels in CentralAmerican countries.

These objectives are achieved by:� Identifying information needs among

stakeholders in the region, which are translatedinto a need for indicators according to issues(forests, land use, natural disasters, and socialdynamics

� Collecting existing data from the region incollaboration with about 50 regional, national,and local institutions

� Preparing and compiling data on a CD-ROM,thus converting available data into feasibleindicators and accessible information.

The rural sustainability indicators project aimsat developing indicators with policy relevance atdifferent tiers of decisionmaking. For this pur-pose, indicators at the regional level (including sixCentral American countries: Costa Rica, ElSalvador, Guatemala, Honduras, Nicaragua, andPanama) have been further developed at thenational and even at the local level in some casestudies.

The design of indicators to monitor sectoralsustainable development requires a systemicapproach, but management and policymakingremain predominantly organized by sector. Forthis reason, the CIAT-World Bank-UNEP ruralsustainability indicators project includes sector-level indicators, in addition to developing

environmental and sustainability indicators atdifferent analytical levels and for different dimen-sions (political/administrative and ecological/environmental). This paper is one of the sectoralcase studies in the larger study; a second study isthe conceptual water-sector case study presentedin Annex B.

Outline

The forest sector paper is organized in five mainchapters:

1. Introduction, including objectives and a briefdescription of the rural sustainability indicatorproject for Central America

2. Key conceptual issues in the development andapplication of forest indicators

3. The regional view for Central America

4. The national level, with examples based ongeo-referenced and tabulated indicators fromthe forest sector in Honduras

5. Further examples of possible developments ofadvanced forest indicators.

Chapter two provides an overview of selectedconceptual issues relevant for forest sector indica-tors. The chapter outlines a number of selectioncriteria, as well as examining the considerationsnecessary for forest indicators from the perspec-tive of what information could be demanded andhow it could be organized, barring limitations inavailable information.

Chapter three shifts to a discussion on indica-tors based on available data. The purpose is topresent national level indicators for all six coun-tries in the region. It is based on data compiled byCIAT and additional contributions from othersources, and does not include desired indicatorsfor which no data are available.

Chapter four is an example of forest sectorindicators at the national level in Honduras. Thischapter is also based on available data; if impor-tant topics are neglected it is due to lack of dataand not a lack of importance or demand. Thepurpose is not to provide a complete descriptionof the Honduran forest sector. Instead this chapteraims to show that even in a decision- and policy-

3Introduction

making environment with limited availableinformation, developing sector indicators canimprove the information level and assistpolicymakers in outlining the current state anddevelopment of the sector. The development ofsector indicator can also facilitate the process ofsetting development goals and identifying infor-mation gaps.

A national sector overview study demonstrateshow different types of information can be pro-

cessed and applied (such as tabulated data,trends, indicators, or geo-referenced data).Indicators are not only a way of convertingexisting data into useful information, but throughthe identification of information needs and gaps,the indicator approach can also providepolicymakers and institutions with a frameworkto organize learning systems, harmonize methods,and improve the use of information.

5

CHAPTER 2

Forest Indicators � Conceptual Issues

Perceptions differ as to what constitutes aforest sector indicator. In this chapter abrief introduction is given to selection

criteria and indicators relevant to the forest sectorin Central America. These are compromises madebetween many different interests, so a commondenominator is difficult to apply. Box 1 presentssome characteristics of the forest sector that needto be taken into consideration when selectingindicators.

Selecting Forest IndicatorsIndicators are a way of dealing with complexdecisions in a decisionmaking environmentcharacterized by both lack of, and unstructured,information. They also offer a low-cost approachto dealing with incomplete information in adecision context with many possible impacts.Rural and environmental indicators can be

instrumental in flagging when and where a policyregulation or other means of intervention isurgent. Another function of environmentalindicators is to monitor the impact of existingpolicies and determine whether they are in linewith policy objectives or whether adjustments areneeded; for example, improved coordination withother sectors.

No consensus exists on exactly what anindicator is or what properties it has. There is atendency to use any available environmental dataas environmental indicators, but this ignores theneed to be selective, to make information out ofdata, and to prioritize. A smaller set of indicatorsis preferable, since decisionmakers are morecapable of dealing with a few well-defined andtested indicators than an incomplete and chang-ing set of data on rural development or theenvironment. Although many potential defini-

Box 1. Forest sector characteristics and indicators

Some of the characteristics of the forest sector call for indicators that capture the dynamics and diversity of theresources and their use. The implications of these characteristics (which are not always exclusively for forests)include:� The need for forest sector indicators to be representative; they need to consider the diversity of forest

resources (including timber resources, food products, global services, species, and soil management).� The large number of different goods and services produced by forests makes it difficult to monitor them all.� The use of forest resources by multiple stakeholders, and within different jurisdictions, exacerbates the

apparent need for lot of information. However, rather than including everything, priorities must be set.� The dynamics of forest resources call for the monitoring of selected indicators consistently over a longer

period of time.� Since natural resources are not confined to administrative boundaries, monitoring is also required across

jurisdictions, such as national boundaries.


tions exist for the format and function of environ-mental indicators, a useful point of departure is todelineate some of the desirable characteristics ofindicators to be considered when selecting a set ofindicators:� Agreement. Definitions (such as “accuracy of

forest cover”) and method of data collection(such as sampling rate) should be agreed onamong stakeholders.

� Consistency. The data should be of a qualitythat makes it possible to compare it bothtemporally (when changes in the naturalresource base occur) and spatially (where thenatural resources are and the changes in theseresources). Special care should be taken withdata and time series that are derived (forexample, a per hectare stock of timber from acertain area that is used to estimate the totalstock) and not measured.

� Policy relevance. The indicators should berelated to the objectives of sustainabledevelopment at the forest management andpolicy level (if these have been defined andagreed upon). This also implies that theindicators should be policy relevant, and thatthe different needs at the regional, national,and local levels (due to different concerns,geographical scales, data availability, andusers) are recognized.

� Clarity. It should be possible for outsiders tounderstand an indicator (for example, termssuch as “forest fragmentation” or“accessibility”). It should also be clear how thechanges and pressures on natural resources canbe explained in both temporal and spatialdimensions.

� Representativity. The indicators shouldrepresent wider and scarce environmentalinformation because of correlation. As suchthey become an alternative to collectingadditional, costly new information.

� Feasibility. An indicator should consist of eitheravailable data or data that can be collected in acost-efficient manner.Failure to identify and apply indicators often

occurs because they are selected on the basis ofwhat different stakeholders and sectors considerrelevant, without consideration of data availabil-ity and monitoring costs of the underlying data orthe overall relevance for decisionmaking. Ifeveryone is asked what indicators may be impor-

tant, the answer will be that everything is impor-tant. This offers no grounds to prioritize whatinformation is the most needed. Because theidentification and development of environmentalindicators is based merely on stakeholders’narrow interest, the indicators become a reflectionof narrowly defined policy objectives, rather thanappropriate decisionmaking tools serving broaderneeds.

Applying these criteria to the process ofselecting indicators results in sets of environmen-tal and rural indicators that provide informationto decisionmakers, the public and private sector,and other stakeholders. For decisionmakers thedevelopment and use of indicators includes threekey questions:� What kind of information is required? Information

needs span from the determination of abaseline to the monitoring of progress.

� How to make best use of existing information?While not all existing information is relevant oravailable at once, the challenge is to haveaccess to and process information to addressmost pertinent needs. (There are drops ofinformation in the ocean of data).

� What are the gaps in the available information?Demand for and availability of information arebalanced to identify the most importantimprovements. Additional information is notalways relevant if costs are considered.

These three questions demonstrate the com-mon problem of demand for and supply ofindicators and data. The next chapter discussesthese aspects in more detail.

Demand and Supply of IndicatorsWho or what decides which indicators are

relevant to monitor a sector? On the one hand,there is demand from decisionmakers and othersfor information relevant to decisionmaking aboutthe forest sector. On the other hand, there arelimitations to the available information (supply),and therefore also to the extent that it is possibleto develop indicators. Is the need for informationor the availability of information deciding whichindicators to include in monitoring?

Figure 1 illustrates this discrepancy betweenthe demand and the supply side of environmentalindicators. The supply side describes the steps,from the selection of available data to describingthe relevant information. The point is that not all

7Forest Indicators — Conceptual Issues

available data is relevant, and the quest is todistill what is reliable and relevant information. Adiscussion about filling possible information gapsis also necessary. The concern is about dataquality, the relevance to policymakers, and thecosts of monitoring the data, as well as otherselection criteria introduced in chapter two. Fromthe available data it is possible to advance to thefeasible indicators by applying additional criteria,such as the possibility of collecting data over time.Finally, information is created on the basis ofwhich decisionmaking is possible.

The demand side pulls in another direction. Itis determined by the stakeholder and sectorinterests, and the need for information reflectswhat may be policy relevant. However, stakehold-ers and sector interests tend to move beyond whatis necessary, because the demand for informationalso reflects the perception that more informationis the path to more influence (as long as the costsof information are not considered). The selectionof policy-relevant indicators should thereforenarrow the wider expressed need for data byaddressing key information needs (setting priori-ties) and balancing these with feasible indicators.

Many examples of the discrepancy betweendemand for and supply of information can befound in the real world, including biodiversityand environmental services provided by forests,which currently attract great interest. This in-creased interest is reflected in the demand for

indicators on these topics. However, the lack ofavailable data makes the supply of indicatorsalmost infeasible (in the Central America case).The information can be collected (such as speciesmonitoring) but at a high cost. In addition theresources may be more efficiently used to preventbiodiversity loss that has already been establishedrather than to continue to measure its decline.

A first step toward solving the dilemma of themismatch between supply and demand can be todevelop policy-relevant indicators from theavailable information, and then to cover identifiedinformation gaps. There are, however, some issuesto be considered regarding an approach that onlyuses available data:� Available information is never complete.

Indicators are representative of a much largerset of data than has been used for thedevelopment of the indicator. The indicatorsare proxies for different types of impacts. Forexample, the reduction in the area of naturalforests is an indicator of possible reduction inbiodiversity. Indicators are a way ofstructuring existing sources of information inorder to monitor the main impacts of complexpolicy decisions.

� Data collection and processing has a cost. Evenexisting data may not be available without acost, and the development of indicators shouldinclude the costs of making both existing andnew information available.

Figure 1. Demand and supply side of environmental indicators

Discrepancybetween what is

feasible (supply) andwhat is wanted

(demand)

Supply side

Existing data

Feasible information

Feasible indicators

Available data

Demand side

Core issues on the agenda of stakeholder

Priority information reflection core issues

Policy relevant indicators (compromisereached between several stakeholders�

core issues)

Possible data reflecting core issues


� Data sources are not always reliable orcomparable. Every source of environmentaldata is in itself an indicator, because data issampled. For example, a forest inventory orbird population count is seldom made on thebasis of a complete assessment. The quality ofdifferent sources of data is variable, whichplaces a constraint on validity.

� The spatial and temporal distribution of dataderived for indicators are dimensions to beaddressed in order to monitor changes andpressure on natural resources in both the local,national, and regional context.

If an approach is selected that focuses onindicators with existing data, another commondilemma for the forest sector becomes apparent:timber production tends to dominate the feasibleindicators, since most available data are oneconomic timber production. In contrast, fewerdata are available on non-timber functions(biodiversity or carbon sequestration). Since thepurpose of a decisionmaking process based onforest sector indicators is to achieve a sustainabledevelopment, goods and services of the forest sectorother than timber production need to be included.Indicators are not an end in themselves, and thedevelopment of indicators for other forest func-tions is not crucial for sustainable development,but indicators can support sustainable forestdevelopment when relevant data are available.

It is possible to widen the scope of the forestindicators to include non-timber functions, but itrequires efforts in both collecting data and devel-oping methods for data collection. The principlebehind using indicators in decisionmaking is toeconomize on data collection. Data on a fewspecies that can represent impacts on forestresources may therefore be sufficient. It should benoted that existing data on forests (for exampleforest cover, forest types, and timber stocks),unlike data for most other sectors, also are quasi-indicators for areas such as biodiversity or carbonstorage. The question is how much misinforma-tion on the sustainable development of forests isgenerated by relying largely on existing informa-tion on forest area and timber production?

Given the various demands on any given set ofindicators — criteria for relevance, the supply anddemand for indicators and the information thatthey contain, problems of missing data and weak

institutions — it is important to clearly identify acore set of indicators. These indicators shouldhave the following characteristics: policy rel-evance, appropriate geographical-level focus, andimplementation feasibility. The following sectionpresents a set of candidate indicators for the forestsector.

Candidate Indicators for the Forest Sector

Indicators in the forest sector have been thesubject of extensive work in many parts of theworld. The aim has been to introduce a verifiablestandard to ensure that criteria for sustainableforest management can be monitored, and eventu-ally also certified. The organization is commonlybased on a two-tier hierarchy:� Policy level: Forest principles and guidelines

� Management level: Forest criteria andperformance indicators.

Although there are some similarities betweenindicator sets to monitor sustainability at theforest management level and the development ofa system of forest sector indicators that capturesthe development of key policy factors at thenational or regional level, there are also funda-mental differences. The monitoring andcertification of sustainable forest management (SFM)and the forest management level is a potentialsource of existing data, but not all of this willconverge at the policy level.

Performance indicators are used to measurecompliance of a defined set of criteria for SFMwithin the overarching forest principles. Whilethis approach has been generally applied, it is aweakness that the principles and criteria areformulated in broad terms that can make develop-ment of indicators cumbersome and reduce itsapplicability. The indicators reflect good inten-tions and common sense, but offer fewopportunities to measure concrete performance orimpact. The challenge is how to delineatesustainability through the agreed criteria and itsapplication; that is, how to make the indicatorsmeasurable and operational.

In sum, stakeholders in the forest sector havedifferent objectives and different needs. The forestindustry and part of the public administrationmay be concerned with the availability of rawmaterial, while others may be concerned with theextent and protection of the remaining natural

9Forest Indicators — Conceptual Issues

forest resources. The differences in policy objec-tives could be resolved with a core set ofindicators that is not targeted at supportingspecial interests only. Indicators may servedifferent policy objec-tives, but the indicatorscannot themselves bepolicy objectives.Finally, there are alsodifferences due theabsence of a frame-work to develop andapply indicators.

A set of indicatorshave been put forwardfor Central America forforest management level, including five criteria and50 indicators.4 These are the “forest principles”developed in the same spirit as other forestindicators at a number of regional meetings (suchas the Helsinki and Montreal processes).

The indicators developed in the criteria andindicators framework can be difficult to apply inpractice. One reason is that they are developed ina context where the indicators reflect policy orstakeholder objectives rather than a tool tomonitor the development or performance of ahigher level of objectives. It is, however, possibleto identify a typology of the different types ofpossible indicators; a generic list of types ofindicators might include:� Basic data (forest area or the annual extraction

of logs)

� Trend data (rate of deforestation) and other datareflecting temporal changes

� Impact data (change in the area of forestaffected by forest fires) where changes arerecorded in a spatial setting

� Check lists (existence of forest legislation)

� Qualitative information (identification of stakeholders)

� Economic impact data (improvement ineconomic profitability or forest products trade)

� Social impact data (accessibility, both physicaland legal, to forest resources).

A simpler, indicator-type framework could becreated consisting of basic data, derived indica-tors, and combined indicators. It is suitable tomake a distinction between different classes of

indicators (derived and combined indicators), andto start out with the basic data that are beingcollected in the field for the purpose of monitor-ing the forest sector or for other reasons (Table 1).

A forest inventory at the national level is thecore of the forest resource basic data. A forestinventory is produced at high cost. Most countrieswith an inventory prepare it at intervals of at leastone decade, but several countries lack a forestinventory. Moreover, there is a large degree ofsampling when an inventory is made, and thestandard deviations on the inventory estimatesprovide an indication of the reliability of data. Aquestion therefore arises as to what extentchanges over the years represent significanttrends or sampling uncertainty. The data onforests are in general estimates, with a proportionof uncertainty especially for non-commercialstands. However, once established, a forest islikely to be in place for a number of years, and itis possible to acquire relatively accurate informa-tion on an annual basis by prolonging informationfrom the previous inventory.

To make environmental indicators applicableto the forest sector (as well as other sectors of bothrural and urban origin) a set of indicators could bedefined following these indicator types. Theindicators in this set could be monitored on aregular basis, even if the frequency differs. Theforest area and timber stock may, for example,only be assessed every decade, while assessmentsof removals may be annually and prices quarterly.Examples of possible indicators are listed in Table2. Hardly any of this information is readilyavailable in Central America, and the cost ofobtaining it should be carefully weighed againstthe benefit for decisionmaking. Collecting infor-mation for indicators has no impact unless thedata is absorbed in policymaking.

Table 1. Types of forest indicators

Type of dataype of dataBasic dataBasic data(absolute)

Derived indicatorsDerived indicators(relative)

Combined indicatorsCombined indicators(absolute)

Explanation Primary datarimary datacollected

Change in basicChange in basicdata over a timedata over a timeperiod

Combination ofCombination oftwo sets of basictwo sets of basicdata

Example Forest areaForest area Change in forestChange in forestarea

Average volumeverage volumeper hectare orProduction perroduction perhectare


Table 2. Some candidates of forest indicators

Baseline data

(the state at a givenlocation or point in time)

Derived indicators

(reflecting spatialdistribution and temporalchange in baseline data)

Combined indicators

(customized frombaseline indicators)

1. Forest resourceavailability

(subdivided to species,age- and site-classes)

Forest areaProductive forest area(e.g. NAI > 1 m3 p.a.).

Area of primary forestArea of legally protectedforest conservationareas.

Area of deforestationArea of reforestationand afforestation

% rate of change inforest area

Forest fragmentationExtent of forest frontierForest area per cap.

2. Volume andproduction

Forest inventoryProduction of industrialroundwood andfuelwood

Inventory of non-timberforest resources.

Annual growthTrends in the annualremoval of timber

Extraction of non-timber forest products

Net annual increment.Removal per ha ofproductive forest.

Balance betweengrowth and removal(potential misleading)

Forest fragmentation

3. Consumption andtrade

Consumption ofindustrial roundwoodand fuelwood

Volume of tradeValue of trade

Change in consumptionpattern

Change in tradepatterns

Change in trade value

Wood products massbalance

Deficit in trade balance($) of wood products

4. Ownership Forest area in public,private and commonownership.

Change in ownershipstructure.

% of area in publicownership

5. Economic impacts Timber pricesCost of timberproduction

Public sector revenuefrom forest sector

Change in timber pricesChange in public sectorrevenue

6. Social impacts Number of peopledepending on forestresources

Conversion of forestareas throughencroachment

7. Biodiversity Species inventory (e.g.number and spread ofendangered species).

Changes in speciesdiversity and density.

Indices of biodiversityrichness

8. Environmentalservices

Stock of carbonForest areas in riparianzones and catchments

Number of forestrecreational visitors

Change in carbon stockLocation of forestsprotection water sheds

9. Forest fires Fire incidents and areaburned

Change in forest firesincidents and areaburned

Loss of timber andrelease of carbon dueto fires

11

CHAPTER 3

Regional Forest Indicators

At a regional level, indicators for theforest sector are supposed to provide ageneral overview of the state and

development of the forest sector in each countrythrough a comparison of key information. A coreset of indicators should be limited, and althoughany topic could be considered important bysomeone, it does not imply it deserves its ownindicator. A set of a few standardized indicatorsfurthermore focus and makes inter-countrycomparisons feasible.

Another argument for focusing on a few,rather than a large set, of regional indicators onforests is that there is no super-national decisionmaking for the forest sector at the regional level.Policy making occurs at the national or locallevel while at the super-national level negotia-tions and collaboration occur that may result innational policy making. There are some efforts toproduce standardized environmental data in

order to respond to international environmentalagreements or organizations collecting suchinformation. There are also collaborative effortsin the region to compile regional data that makesit possible to compare the state of the forestsector and development across national borders.The scope for decision making, however, remainsnational but regional collaboration and informa-tion sharing provides new dimensions fordeveloping and using these indicators.

A set of core indicators at the regional levelcommonly consists of national indicators fromeach country. One of the basic data at the coun-try level and for a regional comparison, that apossible core indicator can be based on, is theforest area (Table 3).

From the basic data the relative forest covercan be estimated. The data from Table 3 arecompared with a set of data compiled by FAOand similar data made available from national

Table 3. Forest area in Central America(hectares)

n.a. : data not availableSource: CCAD 1998: �Estado del ambiente y los recursos naturales en Centroamerica 1998�, Central American Commission ofEnvironment and Development (p.94).

Land area Primary forest Secondary forest PlantationsTotal forest

area

Guatemala 10,889,000 3,030,200 360,000 89,900 3,480,100El Salvador 2,097,000 34,298 275,989 6,593 316,880Honduras 11,249,000 5,989,000 n.a. 8,647 5,998,247Nicaragua 12,142,800 3,700,000 n.a. 64,172 3,764,172Costa Rica 5,113,300 1,305,980 388,000 151,707 1,845,687Panama 7,551,700 2,396,000 n.a. 26,724 2,422,724


forest authorities (Table 4). Neither of the datasets can be proven to be wrong (or correct) andthe differences merely illustrates uncertainty ofdata sampling and different definitions, e.g., offorest cover. The difference in the particular casecan be a result of the definitions used, but sincethe difference is not one-sided there are probablyalso differences in methods applied withincountries. The point is that even when asking thesame question there are several answers that canbe derived.

The inclusion of changes in the forest areaover time, i.e., basic data on the forest area in

different years, can provide some indication ofthe net change in the forest area. A decline in forestarea is interpreted as deforestation (Table 5),although reforestation and afforestation mayactually cover some liquidation of mature timberreserves in natural forests.

In the data from CCAD a somewhat differentapproach has been used based on the assumedarea of annual forest loss rather than single yearestimates of forest cover (Table 6). These data maybe uncertain but they do provide an indication (asany indicator only will do) of the rates of nationalforest losses in Central America.

Source: Data from CCAD and FAO are adapted from CCAD 1998: �Estado ambiente y los recursos naturales enCentroamerica 1998�, Central American Commission for Sustainable Development (p.94-95); Guatemala: PAFG,1996; Honduras: COHDEFOR, 1994; Nicaragua: PAF-NIC, 1992; Costa Rica: MAG-IMN, 1992; Panamá:INRENARE, 1992. (National sources from CIAT).

Table 4. Different sources of relative forest coverCCAD data

%FAO data

%National Sources

%

Guatemala 32.0 35.4 34.4El Salvador 15.1 5.1 n.aHonduras 53.3 36.8 53.2Nicaragua 31.0 45.8 35.41Costa Rica 36.1 24.4 34.7Panama 32.1 37.6 44.47

Table 5. Rate of deforestation � FAO data

Source: FAO,1985, Forest Resources 1980, FAO, Rome and FAO,1995, Evaluacion de losrecursos forestales 1990, FAO, Rome.

Deforestation1980-1985

(area ha) / (rate %)

Deforestation1981-1990

(area ha) / (rate %)

Guatemala 72,000 / 1.6 81,300 / 1.9El Salvador 4,000 / 2.8 3,100 / 2.5Honduras 48,000 / 1.3 111,600 / 2.4Nicaragua 105,000 / 2.3 124,000 / 2.0Costa Rica* 65,000 / 4.0 49,600 / 3.5Panama 36,000 / 0.9 64,400 / 2.0

Table 6. Rate of deforestation � CCAD data(Data from mid 1990s)

Source: Adapted from CCAD 1998: �Estado ambiente y los recursos naturales en Centroamerica 1998�, CentralAmerican Commission for Sustainable Development (p.98).

Forest Area(ha)

Deforestation(ha)

Rate of deforestation(%)

Guatemala 3,480,100 90,000 2.59El Salvador 385,087 11,653 3.03Honduras 4,536,700 108,000 2.38Nicaragua 3,764,172 100,000 2.66Costa Rica 1,845,687 18,000 0.98Panama 2,422,724 51,000 2.11

13Regional Forest Indicators

The uncertainty of the forest area data isincreased when moving towards indicators ofchange in the type of deforestation. A loss in forestarea in any one period can occur as a loss ofnatural and secondary forests less the incrementin forest cover from reforestation of formerforest land and afforestation of other types ofland. The matrix for changes in forest land area(Table 7) includes not only the loss of forestthrough deforestation but also the conversionbetween different forest types and of mature toyoung forest.

The matrix provides additional informationthan just the net deforestation rate, which iscrucial in understanding the impacts of forestchange. The deforestation rate does not capturewhether there is a liquidation of the forestresource towards clearing forest for replanting or

natural succession. It provides more of anoverview of the dynamic changes in the area ofdifferent forest types. The matrix, however, doesnot reveal how large the parts of a forest that areunder influence of logging and possible degrada-tion. The downside of the matrix approach isthat these data are not readily available, but thisapproach can be used to validate data on thearea of deforestation, reforestation and afforesta-tion. To monitor these trends over a longer periodis ambitious and none of the countries in CentralAmerica appear yet to have such detailed levelof information.

A further level of detail of forest indicatorsbeyond the forest area concerns the forest re-source in terms of timber stock, growth andremovals. Basic data on the standing volume aswell as the current net annual increment (NAI)

Table 7. Matrix for changes in forest areas

Notes:1. Net deforestation is the �conversion of forest to non-forest area� minus �non-forest area converted to forest.�2. The conversion from forest area with high stock (older forest) to low stock (young forest) is revealed in the matrix. The young stock may have a higher annual growth than the older forest, but the capital stock is reduced.3. The matrix does not include impacts like low-impact logging in primary natural forests or commercial thinnings in plantations which do not result in a conversion to another forest type. The matrix does not take forest degradation or high grading into account.

Converted toConverted to

�-------------------Converted

From �Primary

natural forestnatural forest Secondary forestSecondary forestPlantation

(reforestation)Plantation

(afforestation)

Forestconverted toconverted to

non-forest landnon-forest land

Forest area andorest area andnon-forest landnon-forest land

converted

Primary naturalrimary naturalforest

Primary naturalrimary naturalforest loggedforest loggedand naturaland naturalsuccession

Logged andLogged andreplanted

Conversion ofConversion ofnatural forestsnatural foreststo non-forestto non-forestuses

= Conversion= Conversionof naturalof naturalprimary forestprimary forestareas

Secondary forestSecondary forest(naturalregrowth)

Logged andLogged andnaturalregrowth

Logged andLogged andreplanted orreplanted or"improvement"of degradedof degradedforest landforest land

Conversion ofConversion ofsecondaryforest to non-forest to non-forest useforest use

= Secondary= Secondaryforest areaforest areaconverted

Plantations(reforestation/afforestation)

Failedplantationstaken over bytaken over bynaturalregrowth.

Replanting aftereplanting aftercompletedrotation.

Conversion ofConversion ofplantations toplantations tonon-forest usenon-forest use

= Plantations= Plantationsconverted

Non-forest areaNon-forest area Natural forestNatural forestgrowth ongrowth onabandoned landabandoned land

Plantations onPlantations onnon-forest landnon-forest land

= Non-forest= Non-forestareaconverted toconverted toforest

Increment ofIncrement ofarea with youngarea with youngforest.

= None= None(Increment in(Increment inarea of primaryarea of primarynatural forestnatural forestnot possible)not possible)

= New= Newsecondaryforest

= R= Reforestation = Afforestation= Afforestation = Conversion= Conversionof forest toof forest tonon-forestarea

Net forest areaNet forest areaunder someunder someconversion(impacts)


are costly to obtain, and even at the forestmanagement level a comprehensive inventory isexpensive. Only when the stands approacheconomic maturity are there incentives for forestmanagers to intensify the level of inventory. Thereality at country level is that only few countrieshave good inventories of even the commercialvolume of standing timber resources, and foralmost any assessment of the total increment involume this is estimated by a fixed factor derivedfrom experience and permanent measurementplots on the production of m3 per hectare. Whilethe total timber stock and its increment may beindicators in demand it is one example where thecost of the efforts to obtain such informationmay exceed the benefits of more accurate infor-mation.

A feasible approach to the assessment of theuse of forest resources is through basic data ofremovals. This is a pressure indicator that can bemore readily available than the state indicators ofstock and increment. The removal of industrialroundwood in public forests will mostly beknown, e.g., from taxation of concessionaires. Theremovals, however, are basic data that does notreveal whether the resource is sustainably used(e.g., whether the timber stock is non-declining).For the Central American countries estimates ofthe removal of wood is available on a regionallevel including the rate of fuelwood productioncompared to total wood consumption (Table 8).

Table 8 reveals that the major extraction ofwood in the Central American region is forfuelwood. The share of industrial wood produc-tion is rather small in both absolute and relativeterms. These figures do not reveal whether there

is a sufficient and sustainable supply of fuelwoodor whether a reduced fuelwood production couldenhance the supply of industrial roundwood andprocessing of industrial wood leading to furtherdevelopment of a wood based industry.

FAO data are available from 1980 and 1990on the consumption, production and trade inindustrial roundwood and forest products for thesix Central Americans countries.5 This type ofdata makes it possible to develop national woodbalances (a topic further explored in chapter 5).It can reveal to what extent there is a balance inwood consumption and production or whether anation relies on imports. Dependency on woodand forest products imports does not have to be anegative impact in neither an economic nor anenvironmental sense, because trade may assistcountries in utilizing their comparative advan-tages with regard to the costs of production,resource availability, environmental protection,and land-use for other production.

In a regional setting trade is a crucial indica-tor since this is what is linking forest sectors indifferent countries under the given conditions ofmarket access and level of protective measuresor other market interventions. One indicator forthe development in markets is the market priceon industrial roundwood but also on variousother forest products (Table 9).

With a low level of the resource base forindustrial roundwood the forest industry inCentral America is not very developed, and thesetopics may not be exactly where there is a largeneed for further of information on the forestsector.

Table 8. Production of fuelwood and industrial roundwood(reference year is mid-1990s)

Source: Adapted from CCAD 1998: �Estado ambiente y los recursos naturales en Centroamerica 1998�, CentralAmerican Commission for Sustainable Development (p.99).

Fuelwood(million m3)

Industrial wood(million m3)

Share of fuelwood oftotal wood

(%)

Guatemala 7.11 0.31 96El Salvador 5.30 0.32 94Honduras 7.50 0.70 91Nicaragua 3.70 0.27 93Costa Rica 1.69 0.68 71Panama 0.93 0.05 95

15Regional Forest Indicators

Having discussed some of the most common,and important, indicators of the forest sector aswell as the data availability for these indicators,it is now time to focus on the set of regional coreindicators for the sector.

A Regional Core Set of Indicators

One proposition is that the core set of environ-mental data is a selection of basic data. These arethe building blocks for indicators and with anexcellent core set of basic data available in timeseries and spatially, it is possible: (i) to developcustomized indicators to meet the needs of agiven decision process, and (ii) to define andimprove the input to a set of more advancedforest sector indicators which, for example, canbe used for inter-country comparisons and as a

consistent reporting mechanism to internationalagencies and environmental agreements.

Occasionally, the available forest sector dataused as indicators are in a form where the basicdata cannot easily be assessed. This is for ex-ample the case with data on percentage rate ofdeforestation or trade in forest products. Itbecomes a problem for the quality assurance andthe merger of different data sources to build timeseries, and without information about the basicdata a comparison across countries and time isdeceptive.

However, considering that indicators aresupposed to give a first indication of the condi-tion of a sector, creating a core set based onindicators, rather than basic data, can providemore information with less indicators. A core setof regional forest sector indicators can still be

Table 9. Trade in wood based forest products (1993)

Source: FAO Forest Products Yearbook (1993).

Exports(US $)

Imports(US $)

Trade deficit(US $)

Guatemala 3,634 66,913 63,279El Salvador 104 45,121 45,017Honduras 29,306 39,913 10,607Nicaragua 523 6,115 5,592Costa Rica 2,454 111,408 108,954Panama 3,658 37,514 33,856

Table 10. Suggested set of feasible regional core indicators

Notes: All data are from the mid 1990s.1. One of the basic indicators from the forest inventory. The relative forest area can be derived by comparing forest area and total land area.2. The rate of deforestation can be derived from net changes in the forest area.3. The annual wood removal is an indicator of the use of the forest resource (pressure) as well as economic availability and demand.4. Fuelwood share is a derived indicator that reflects to the policy makers the relative importance of fuel wood production.5. The trade deficit in forest products is derived from the export and import values which are derived data from trade statistics.

Forest areaorest area1 2 3

3

4 5

(1,000 ha)(1,000 ha)

Rate ofRate ofdeforestation

(%)

Annual woodAnnual woodremoval

(million m )(million m )

Fuelwood shareuelwood shareof removalsof removals

(%)

Tradedeficit(USD)

Costa RicaCosta Rica 3,480 2.59 7.4 96 63,279

El SalvadorEl Salvador 385 3.03 5.6 94 45,017

Guatemala 4,537 2.38 8.2 91 10,607

Honduras 3,764 2.66 4.0 93 5,592

Nicaragua 1,846 0.98 2.4 78 108,954

Panama 2,423 2.11 1.0 95 33,856


based on the available information. Table 10presents a feasible but not necessarily the onlypossible set of core indicators for the forest sectorat the regional level in Central America. Ratherthan including every single indicator possible atthe regional level the selection of a core set ofindicators reflects a prioritization of the impor-tant indicators. The importance of an indicator ishere both determined by the availability of data(supply side) and policy relevance (demandside).

The core set presented in Table 10 is a prag-matic set of indicators aiming at feasibleindicators that provide a first impression of theforest sector. There is some scope for customizingindicators by a comparison with other basic data,e.g., relative forest cover, forest area per capita,and wood consumption per capita. There is a

bias towards wood production and commercialuse but this is a result of the available data andnot the importance of “soft” issues such asecological, social and institutional issues. Theexplanation is that it is difficult to derive indica-tors on these topics. In the short-term, thedecision makers therefore have to: 1) identifytrade-offs and not use indicators as the only basisfor decisions, and 2) recognize that due to jointproduction in forestry, a production orientedindicator may also comprise “soft” information.In the long-term, however, a strategy for includ-ing and monitoring “soft” issues should bediscussed and implemented.

To be able to discuss forest indicators in moredetail, including indicators that are geo-refer-enced, the next chapter focuses on Honduras andits available data on the forest sector.

17

CHAPTER 4

The Forest Sector in Honduras

Key Information on the Forest Sector inHonduras

Honduras has one of the most developed forestsectors of the countries in Central America interms of economic development and industrial-ization. One explanation is that unlike othercountries in the region a large share of the forestresource in Honduras consists of native pineforests (softwood). These pine forests are locatedparticularly in the high-lands in the center of thecountry while the hardwood forests are predomi-nantly located on the Northeastern part.

The total forest area in Honduras accordingto the ITTO data is 6,112,000 ha with 53 percentin production forest and 47 percent in protectionforest, and these are almost equally dividedbetween the hardwood and conifer forest area(Table 11). The official 1997 statistics fromCOHDEFOR, the national forest agency inHonduras, has a total forest area of 5,989,600 ha.It is a 2 percent reduction equal to the approxi-mate deforestation rate.

The standing volume of the forests is alwaysan uncertain estimation. Figures fromCOHDEFOR suggest that the total volume is 603

million m3 and this is probably commercial stockonly (about 100 m3 per ha on average). Of this 36percent is in conifer stands, 58 percent is inhardwood stands, and the remaining 6 percent inother stands like mangroves. However, morethan 90 percent of the timber production (re-movals) is from the pine forests which reflects alikely higher production capacity of this foresttype (lower average stock but higher averageannual increment). The uncertainty and techni-cal difficulties in estimating these figures implythat time series on the development of the timberstock are not likely to be available.

The only data that are available as a timeseries cover a relative short period and only theremoval of industrial roundwood (Table 12). Nodata on fuelwood extraction are available. Notefrom the previous chapter that the industrialroundwood production is estimated to be onlyabout 10 percent of the total wood extraction.Thus the removals of industrial roundwood isonly a small share of the total production ofwood. The largest share is fuelwood with anestimated volume of 6–7.5 million m3 harvestedannually.

The estimation of anexact figure of fuelwoodharvests is highly uncertainbecause the amount offuelwood is not monitored tothe same extent as industrialroundwood removals. The

Table 11. Forests areas in Honduras, 1996

Source: ITTO country profile of Honduras (www.itto.or.jp/forest_update/v8n2/22.html).

Production forests(ha)

Protection forest(ha)

Hardwood 1,932,000 1,357,000Conifers (softwood) 1,308,000 1,515,000Total 3,240,000 2,872,000


removal of industrial roundwood is easier tomonitor because some accounts are kept of thesevolumes, e.g., for taxation purposes. The removalof hardwood, as it appears in the statistics inHonduras, is insignificant but increasing interms of volume extracted, and the bulk of theindustrial wood removal is pine. Annual varia-tions in the production may indicate resourceavailability. However, there is no clear indicationwhether a declining or increasing trend inremovals indicates resource depletion. Thelinkage between the level of removals andsustainable timber production is not straightfor-ward even if harvest is compared to increment.Variations in removals also reflect changes inmarket prices, industrial capacity and institu-tional arrangements, e.g., incentives anddisincentives in the forest legislation.

A general conclusion is that the availability ofdata that may qualify as core indicators for theforest sector are scarce. There are few dataavailable that for example enables the construc-tion of reliable time series. The challenge is to beable to move beyond data extracted from scat-tered sources of information towards a morestrategic approach where the collection of datafor core indicators is balanced with the informa-tion needs of the policy makers in the forestsector. One remarkable observation from Hondu-ras, in particular, is the extensive data collectionpresented in annual reporting performed by thenational forest agency (COHDEFOR). A valuableassistance to COHDEFOR would be to convert

the collected data into a core set of forest indicatorsthat are manageable and have policy relevance.

Developing Geo-referenced ForestIndicators for Honduras

Forest data from Honduras are also available ata sub-national level from different provinces.Moreover, some geo-referenced information isalready available which can be linked to existingbasic data. In this paper, three examples of basemaps are discussed that provide an overview ofthe forest area development in Honduras over adecade and give some indication of the forestfrontiers. These types of indicators are ex postand do not yield direct up-front information fordecision making in the future. In addition, threethematic maps are introduced which depictissues related to forests such as forest fires,floods, and accessibility.6

The spatial presentation of forest indicatorsadds a new dimension to the identification ofnational forest planning compared to otherformats like tables and charts since maps pro-vide an opportunity to present geo-referencedinformation in a superior way, e.g., overlaying ofindicators. Maps are thus a strong form ofpresenting spatial relation and in particular thecorrelation of more than one set of basic indica-tors. However, the major advantages of apresentation of forest indicators on maps canalso be counted among their major drawbacks.One drawback is that maps can make it difficultto assess the quality of the underlying datasources and basic data used, and colorful maps

Table 12. Removals of industrial roundwood in Honduras (1993-97)

Source: Data from COHDEFOR 1997: �Annuario estadistico forestal 1997� as compiled by CIAT.

Conifers(softwood)(1,000 m3)

Hardwoods(1,000 m3)

Total industrialroundwood removal

(1,000 m3)

1997 663.8 41.2 705.0

1996 613.0 32.2 645.2

1995 469.8 4.8 474.6

1994 693.0 3.8 698.9

1993 584.5 5.6 590.1

19The Forest Sector in Honduras

can be deceptive and confirm unreliable data. Acritical factor is therefore the validity of dataand the risk that mapping becomes an implicitvalidation of data sources because a qualityassessment is not feasible from the maps and thecomparability of different sources of informationand definitions is not disclosed. There are, forexample, forest locations in the 1994 map thatare shown as deforested in the 1985 map thatcannot only be a result of afforestation (for themaps, see Annex A). The example is a case ofconifers and mixed forests in the Central andWestern part of the country, and it reveals someof the limitations of maps to describe impacts, ifthe primary data are not consistent with theobjective.

One of the main purposes with the base mapsshown in Annex A is to provide an analyticalbackground for the assessment of more specificissues depicted in thematic maps. An example ofa thematic map is forest area and accessibility tomarkets, which combines several layers ofinformation in one map to address the issue ofaccessibility and pressure on the forest resource(Annex A). Accessibility is defined as a geo-referenced indicator of the time it takes to getfrom a given location to any smaller or largerurban community. The accessibility takes intoaccount road access and topography and to alesser extent mode of transportation. The indica-tor of access to a market or smaller urbancommunity from a given site is also an indicatorof access from a market or smaller urban commu-nity to a given site. In economic terms, access isof importance when determining the land valueof alternative land uses. Hence, access risksexerting pressure on natural forest resources intwo ways: (1) through an economic pressure toexploit the resource, and (2) by augmenting thevalue of land for alternatives uses such asagricultural crop production. A substantial shareof global natural forests is protected becauseexploitation is not economically profitable.

The map reveals (not surprisingly) that noforest cover results in a higher accessibility. Italso shows how access increases along the forestfrontier. What cannot be deducted explicitlyfrom the maps is whether, for example, access

results in deforestation, deforestation improvesthe condition for access or deforestation in-creases the population density and thus thedemand for infrastructure? But it is the experi-ence that improved access also increases thepressure on natural resources. What is oftenhappening with indicators is that they are usedfor deductive arguments, however, in most casesthere are no causal relationships from theenvironmental data.

The second thematic map shows forests andfloods in Honduras. The floods resulting fromhurricane Mitch (in 1998) are here presentedtogether with the forest area. Floods did notoccur in forested areas, but this is no surprisesince forests remain on higher elevations whilefloods occur at a lower. The spatial overviewdoes not lead to clear conclusions on the impor-tance of forest cover to protect against floods.But it does show where there is a possible inter-action between forests and mitigation of floodimpacts worth a further assessment. It wouldalso be expected that in extreme situations likewith hurricane Mitch, the forest cover has onlysmall impacts on floods, but may play a protec-tive role in the case of the more significantproblem of mudslides.

The incidence of forest fires in 1993, if de-picted in a map (Map 6 in Annex A), reveals twointeresting observations. Firstly, the fire inci-dences are scattered equally over a largeproportion of the country except in the forestareas and this could indicate a strong humaninfluence. Secondly, there is a strong concentra-tion on the same Western part of the broadleavedforest core, where the forest area maps revealedfragmentation. Thus from a policy perspective isseems that action could be focused on the fireincidences along this forest frontier.

In the development of forest sector indicatorsthe proposed procedure is to extract informationfrom available data, and from there decide on acore set of indicators. The selection of coreindicators is balanced by data availability,stakeholders’ interests, and policy relevance. It isnot possible to monitor and collect data oneverything, so the development of indicatorsshould be pragmatic in order to be applicable.


Once a core set of indicators for the forest sectoris established it will be available as a tool whichmay be developed to more advanced forest

sector indicators. The following chapter intro-duces three different options of advanced forestsector indicators.

21

CHAPTER 5

Possible Developments of Advanced ForestSector Indicators

Although not feasible at the moment, anobjective may be a more strategicapproach to the development and use of

indicators in the forest sector in Central America.A large challenge is therefore to develop indica-tors that reflect a wider range of forest goods andservices including those where few data areavailable, like social and environmental impacts.As a first step towards this more strategic ap-proach, this chapter is discussing some prospectsfor advanced forest sector indicators.

With the availability of basic data from pri-mary data collection on forests it is possible todevelop more advanced and complex indicators.Three different types of indicators are consideredwhich can be further developed:� A wood utilization balance to determine the

domestic consumption and the sources to meetthis demand. This could give an indication ofdomestic production compared to domesticsupply and the vulnerability there could befrom a dependence on trade in regional orglobal forest products markets.

� A carbon account for the forest sectormonitoring the trend in the net sequestration ofcarbon. This could be of relevance if thenational government wish to link the forestsector in the national policies to eliminate thenet emission of gasses contributing to climatechange.

� An asset valuation of the forest resource whichis an economic assessment of the capital value

of forest resources. This could be included inan expanded version of the national accountsthat takes into account natural capital and itsdepreciation.

There is at present neither sufficient data nornational capacity to develop these indicators forthe countries in Central America. But if thecollection of basic data is systematized andcarried out on a regular basis, there will be apossibility to use this information to developmore advanced forest indicators. This chapterserves as an illustration of such prospects.

A Wood Utilization Balance

Much concern arises around the topic of self-sufficiency of wood and forest products. Oneissue is the maintenance of a sustainable supply offuelwood, fiber and logs from the forest sector,and another is the reliance on supply of forestproducts through trade. An indicator, such as thewood utilization balance, is related to the con-sumption pattern of wood and forest productswhere the supply of roundwood to domesticprocessing and final consumption plus trade isbalanced with the net import of wood and forestproducts.

Production, consumption and trade data forthe six Central American countries for all maincategories of roundwood and forest products areavailable from FAO. These data form the basis fora wood utilization balance. For each forest prod-


uct commodity the consumption is calculated byadding imports to production minus exports(consumption = production + imports – exports).One example is for sawnwood in Honduras forwhich data are presented in Table 13.

The sawnwood data show that the productionin Honduras has dropped by one third. Further-more, due to this drop in production, reducedexports or structural changes in the sawnwoodindustry have occurred. (There may be an expla-nation, e.g., a trade regulation of sawnwood). Thedomestic consumption has increased by 6 percentfrom 1980 to 1994.

The main categories of final consumption offorest products are listed in Table 14. Industrialroundwood or different types of wood pulp arenot listed to avoid double counting, because theseare intermediary products, i.e., roundwood isused for sawnwood and panels, and pulp is usedfor paper and paperboard.

With an assumption of conversion factors toproduce forest products it is possible to estimatethe total consumption of wood. If it, for example,is crudely assumed that the average conversionfactor is 50 percent for all forest products, the totalconsumption of wood in forest products is 688,000CUM in 1980 and 830,000 CUM in 1994. Theconsumption of industrial roundwood in Hondu-ras was 1,082,000 CUM in 1980 and 540,000 CUMin 1994. However, this is not a relevant compari-son, because the Central American countriesimport most of the paper and paperboard con-sumption. The reason for this is that the forestsector in any of the countries is not sufficiently

large to supply a pulp and paper industry. Thetotal consumption of wood could be in the orderof 6.5 million CUM in 1994. With a population ofaround 5.8 million this is about 1.13 CUM percapita, with almost 90 percent consumed asfuelwood.

The preparation of an index for wood con-sumption is feasible but rely on various datasources, the technical conversion rates, and theproduction and net trade. In the estimation thereis a risk of double counting, e.g., residues fromsawmills are used as fuel or as an input to paperor fiber board production and it is a challenge tofactor these out. As a control, the wood balancecan also be addressed from the supply side.

Carbon Accounts

One of the most direct examples of joint produc-tion is that between timber and carbonsequestration. A distinction can be made betweenthe rate of carbon accumulation and the storagecapacity of the forests, where the important is thelatter, which is a functional relationship betweenthe rate of sequestration and the time lag beforerelease and possibly replacement through re-growth. The time lags and the overlappinguptakes are what in the end results in a netsequestration of carbon in a plantation.

Estimating the carbon storage in forests is notinteresting because carbon provides direct ben-efits in itself to the forest owner, localcommunities or the country. The reason is thesuggested mechanisms of the Kyoto protocol inthe Framework Convention for Climate Change(FCCC) where there may be a possibility of a

Table 14. Final consumption of forest products in Honduras1980 1994

Fuelwood (1,000 CUM) 3,704 5,701Sawnwood (1,000 CUM) 291 309Wood based panels (1,000 CUM) 10 17Paper and paperboard (1,000 MT) 43 89

Table 13. Sawnwood production and consumption in Honduras

Year 1980 1994

Production 560 362+ Imports 0 0- Exports 269 53= Consumption 291 309

23Possible Developments of Advanced Forest Sector Indicators

future market for carbon sequestration or reducedreduction in commitments to net reductions.

In the forest sector carbon accounts are madeby linking them with the forest inventory. If therewas a general interest in the amount of carbon inthe forest sector the standing volume would be animportant indicator, although there are differencesin carbon content among different tree species andamong different forest ecosystems when litter andsubsoil organic matter are included. The assess-ment of the current carbon stock is commonlyassessed by multiplying the following factors:� The volume of commercial timber stock

according to national forest inventory.

� The carbon content per CUM (dry-matterweight, which varies among tree species, times0.5, the carbon content of dry-matter).

� The “expansion” factor, which takes intoaccount non-commercial above ground non-commercial wood biomass, debris, litter andbelow ground biomass in roots.

The carbon content will thus be estimatedbased on the inventory and technical coefficients.A full measurement of the carbon content isunlikely, although some trials could be made toestablish a more concise basis for the content ofcarbon in wood biomass of different tree speciesand the relation between commercial timber in theinventory and total biomass in forests includingsoils.

The change in the carbon stock can thereafterbe monitored by comparing stocks in variousyears. The bottleneck is the lack of a sampledinventory of commercial timber stocks. Althoughdata exists on the forest stock these are in the caseof Honduras not derived through sampled datacollection but through the application of fixedcoefficients to the forest area of different foresttypes. The carbon account will then vary only bythe variation in the forest area, which would makeit a quasi-indicator.

Asset Values of Forests

Forests are a natural capital asset. As a renewablenatural resource, they also offer an opportunity toinvest in natural capital through afforestation of

non-forest land, reforestation of former forestland, and rehabilitation of degraded forest land.The natural capital of forests can be included, forexample, in the asset balance of the nationalaccounts, and the asset value can be an indicatorof the development of the forest capital in ameasure of national wealth. It is not a rule thatforest capital shall be non-declining. The point isthat excessive extraction results in a decliningasset value, which in turn becomes a loss ofwelfare. This loss in welfare can be monitored,and preferably counterbalanced by an increase inwelfare somewhere else.

The valuation of a forest is unfortunately notstraightforward even if only the marketed (price)products are considered. One simple approachhas been to multiply the standing volume with aunit stumpage value, i.e., the difference betweenthe market price and costs of extraction. Althoughthis approach seems straightforward it is notrecommended because the value of a non-matureforest stand exceeds the current realization value,i.e., the stumpage value (Figure 2). The asset value(land expectation value) is the net present value ofall future returns from the forest and this requiresin addition to information on prices also informa-tion on harvesting schedules and loggingvolumes, future prices and the discount rate.Furthermore, in addition to the value of timber,the forest land also has a value due to the eco-nomic return of growing trees and an opportunitycost from excluding other uses.

Figure 2. The land expectation-, land-, and stump-age value of a forest stand


The asset value of the timber stock and forestland is an economic approach to sustainability,but it does offer some useful approaches to theassessment of the development of forest resourceswhich is not considered in a physical account ofhectares and cubic meters. This is, for example,the ability to consider a forest as natural capitaltaking into consideration the location and qualityof the resource. It is a high-end indicator thatcombines basic and derived indicators of bothphysical and economic dimensions. An exampleof a possible application is for the inclusion ofnatural resources in the expanded system ofnational accounts. Few, if any, countries havemade it all the way to an asset valuation of forestresources for this purpose.

The World Bank has estimated natural capital,including forest assets, for the measure of nationalwealth. Table 15 provides an example of countrylevel data for Central America. These are indica-tors at the higher end of more advancedindicators, composed by basic informationtogether with technical and economic assump-tions. The final product can be useful informationfor decision makers at the national level. Thedownside is that these indicators can not easily beverified or reproduced by the national level users,who may therefore not have the same trust inthese figures compared to indicators developedby the users.

Naturalcapital

Pastureland Crop land

Timberresources

Non-timberforest

benefitsProtected

areasSub-soilassets

$ percapita Per cent of natural capital

Costa Rica 7,860 19 72 2 1 5 ..

El Salvador 1,150 22 77 1 0 0 ..Guatemala 1,720 18 54 10 6 9 4

Honduras 3,380 12 47 24 6 7 3

Nicaragua 3,690 15 57 16 10 2 0

Panama 6,300 15 63 4 5 13 ..

Source: World Bank, 1997: �Expanding the measure of wealth - indicators of environmentally sustainable development�, Environmen-tally Sustainable Development Studies and Monographs No. 17.

Table 15. Country level natural capital estimates in Central America

25

CHAPTER 6

Concluding Remarks

This paper is an attempt to derive a core setof indicators that would inspire the use ofindicators in the forest sector in Central

America. The approach used in the paper makes itclear that by working with feasible information itis possible to make a first leap without loosingtouch with neither the users’ needs nor the realityof the data. A major drawback is that existing datareflect economic uses and there is little informa-tion available to meet the needs for indicators onthe environmental and social functions of theforests.

A fundamental approach towards the develop-ment and use of indicators is suggested in thepaper. It consists of the following steps:� Identify available data

� Prioritize the data by selecting the feasible setof core indicators

� Develop composite use of indicators by usingthe core set of indicators to develop policyoriented indicators.

� Apply both the core set of indicators and thecomposite indicators to monitor the resource,to develop forest policy, and for a dialoguewith stakeholders.

� Extract the policy relevant information fromthe results of the monitoring.

Not all available data should be considered asindicators for the forest sector, and not all desir-able indicators can be supported by sufficient data

(within reasonable costs). The discrepancy be-tween desired and feasible indicators imply thatthere is insufficient possibility (technically andeconomically) to support a wide range of indica-tors. There is a need to prioritize and developindicators within the scope of forest sector policy,existing learning systems, and technical skillsamong the users.

Traditionally, there has been a narrow focus onthe timber production side in discussions of theforest sector. It should be stressed that there areother forest resources and services than timberproduction, and some of these can also be inven-toried and included as forest indicators. Oneexample is the monitoring of key species of bothplants and animals, which reflects the disturbanceand development of the diversity of the forestecosystem. In the case of the Central Americancountries this may be desirable from the policypoint of view but not yet feasible from the datapoint of view.

Non-timber forest indicators are anotherbranch of forest indicators that can be developedlater, but just because these resources are notcurrently being as closely monitored or mentioneddoes not imply they lack importance. The purposeof this paper has been to use what is feasible fromavailable data. There are some good reasons whythe monitoring of the timber resources is anapplicable point of departure:� The information on the timber resource is more

readily available due to the commercial


importance although data quality may vary. Anational forest inventory is the outset for a coreset of indicators for the forest sector.

� The joint production in forests indicates that ifan area is covered with forests (and inparticular also if this forest maintains somestructure and biomass) this is an indicator of awider existence of other forest resources thantimber. There are modifications to thisassertion, e.g., when a natural forest is replacedby a plantation.

It is a finding that the ability of a country todevelop the core set of indicators can be judgedon the quality and frequency of the national forestinventory.

The lesson for decision makers and otherstakeholders in the forest sector is that the sectorindicators are only one part of the availableinformation for decision support and communica-tion. Indicators can be developed where this isfeasible but their outcome needs to be combinedwith more soft approaches to topics that aredifficult to monitor:� Social issues and impacts

� Ecological services and biodiversity

� Institutional capacity and changes in policyobjectives.

These may at some stage also be reflectedthrough indicators but that stage has not yet beenreached for the forest sector in the Central Ameri-can countries. The development and use ofindicators is complementary to gaining furtherinsight and understanding of social, ecologicaland institutional processes.

It should not be neglected that information(objective or manipulated) in the form of indica-tors can be powerful in providing access to thepolicy agenda. If there is sufficient information toraise an issue in the forest sector it can, throughpublic attention, open the policy agenda for forestpolicy changes. Finally, indicators are not to beready-made by technical experts for any predict-able occasion. Indicator development can havemore impact if an agency (or stakeholder) basedon sector knowledge and the building blocks froma core set of indicators can develop their owncustomized set of indicators describing a particu-lar policy problem. Indicators should not drivethe policy agenda but they should be instrumentalin assisting stakeholders including decisionmakers in defining policy problems and solutionsin the forest sector and how to monitor theirimpacts.

27

ANNEX A

Maps of Forest Area and Related Issuesin Honduras7

Base maps: Forest resource maps of the forest area

Map 1. Forest area in 1994 (including protected areas), Honduras

Broadleaved Forest

Coniferous Forest

Mixed Forest

Protected Areas

Source: COHDEFOR, 1996; WCMC, 1996

Broadleaved Forest

Coniferous Forest

Mixed Forest

Protected Areas

Source: COHDEFOR, 1996; WCMC, 1996


Map 2. Forest area in 1985, Honduras

Map 3. Forest area that was already lost in 1985, Honduras

Broadleaved Forest

Coniferous Forest

Source: OEA, 1984

Broadleaved Forest

Coniferous Forest

Source: OEA, 1984

Deforested Broadleaved Forest

Deforested Coniferous ForestSource: OEA, 1984

Deforested Broadleaved Forest

Deforested Coniferous ForestSource: OEA, 1984

29Maps of Forest Area and Related Issues in Honduras

Thematic maps: accessibility, floods and fires

Map 4. Forest accessibility, Honduras

Map 5. Forest and floods, Honduras

Coniferous ForestBroadleaved Forest

Mixed Forest

Accessibility

1-2 hours

0-10 minutes10-20

2-4 hours

40-60

8 hours +4-8 hours

20-40

Source :CIAT, 1999; COHDEFOR, 1996

Coniferous Forest

Broadleaved Forest

Mixed Forest

Accessibility

1-2 hours

0-10 minutes

10-20

2-4 hours

40-60

8 hours +4-8 hours

20-40

Source :CIAT, 1999; COHDEFOR, 1996

Broadleaved Forest

Coniferous Forest

Mixed Forest

Partially Flooded Areas

Totally Flooded Areas

Source: COHDEFOR, 1996; SAG /DENDRON, 1998

Broadleaved Forest

Coniferous Forest

Mixed Forest

Partially Flooded Areas

Totally Flooded Areas

Source: COHDEFOR, 1996; SAG /DENDRON, 1998


Broadleaved Forest

Coniferous Forest

Mixed Forest

Fire Incidents 1993

Map Source: COHDEFOR, 1996; ESA/ESRIN, 1997.

Broadleaved Forest

Coniferous Forest

Mixed Forest

Fire Incidents 1993

Map Source: COHDEFOR, 1996; ESA/ESRIN, 1997.

Map 6. Forest fires, Honduras

31

ANNEX B

Conceptual Framework for the Developmentand Use of Water Indicators

Introduction

Sustainable development, and indicators tomonitor it, implies a systemic approach. Never-theless the management and policies for someresources are still organized at the sectoral level.For this reason, in the context of the CIAT/WorldBank/UNEP rural sustainability indicators project—besides the development of environmental andsustainability indicators at different scales andlevels—a goal is to develop sectoral indicators.

Water is one case where the management of aresource is still sectoral. Water resources play avital role in the lives of all people and in thefunctions of ecosystems. At the same time, theymay appear as one of the most endangerednatural resources in the world. The presentdocument is a technical note that introduces aframework that can be used in the developmentand use of water indicators. Two different indica-tor approaches are presented—a project-basedapproach and the Pressure-State-Impact-Response approach.

First steps

The two first steps for any indicator developer arei) problem identification and ii) establishment ofobjectives. The phase of problem identification isnecessary to focus the selection of indicators. Evenif the choice to study water as a sector is a way tonarrow down the selection, an identification of thespecific issues – aspects of water quality or waterquantity – is usually helpful. The problem identi-

fication can be achieved in two ways. One ispublic consultations. Key stakeholders that shouldbe consulted include those who cause or areaffected by environmental problems (e.g., indus-try and population in the project area), those whohave relevant information and expertise (e.g., thescientific community, international and localNGOs, and the media), and those who controlimplementation instruments (governments). Theother way is through literature reviews. Usually acombination of the two is the most efficient.

Both the problem identification and theestablishment of objectives benefit from being asdetailed as possible since that will enable theindicator selection and policy formulation to bespecific. Vague or overly broad objectives, such as“improving water quality”, are of little use inselecting indicators (and may well indicate thatthe project or component itself is not very wellthought out). The establishment of objectives is,however, only the first step—they are not of muchuse if the advances of the project are not mea-sured against them. This implies a need tomeasure the environmental problem at threepoints in time: before the project begins (to obtainbaseline values), during project implementation(to enable changes in the project design if devia-tions from the objectives are noticeable), and afterthe project has ended (to compare baseline valuesto targets). Hence, to identify and establish

Prepared byManuel Winograd, Marta Aguilar, Andrew Farrow (CIAT),and Lisa Segnestam (World Bank)


appropriate and reachable objectives, causal linksmust be clearly identified.

Selection Criteria8

There is no universal set of indicators that isequally applicable in all cases. However, thefollowing criteria are appropriate to most indica-tor selections:

Direct relevance to project objectives. Theindicator selection must be closely linked toproject objectives and the environmental problemsbeing addressed. Where the environmental impactis not the primary objective, an EnvironmentalAssessment process can outline the potentialimpacts and hence help select indicators.

Limitation in number. A small set of well-chosen indicators tends to be the most effectiveapproach.

Clarity in design. It is important that theselected indicators are defined clearly in order toavoid confusion in their development or interpre-tation.

Realistic collection or development costs.Indicators must be practical and realistic, andtheir cost of collection and development thereforeneed to be considered. This may lead to trade-offsbetween the information content of variousindicators and the cost of collecting them.

High quality and reliability. Indicators, andthe information they provide, are only as good asthe data from which they are derived. If the“ideal” indicator to measure a problem is basedon unreliable data, it is common to depart fromthe “ideal” indicator and use proxies instead.

Appropriate spatial and temporal scale.Careful thought should be given to the appropri-ate spatial and temporal scale of indicators. Sincethe environmental impact of project activitiesseldom coincides with administrative boundaries,indicators often need to be measured on differentscales. There might also be lags in time beforeproject effects are felt.

The interpretation of the indicators selectedbased on the above criteria, is further facilitated ifthey are structured according to a framework. Thefollowing section introduces two possible ap-proaches to such frameworks.

Indicator Approaches

A project based approach

One approach is depicted in Figure 3. In this, thefocus is on a project—a water project has bothoverall project objectives it is designed to meet andcomponents by which the implementation of theproject proceeds. The implementation of thesecomponents, combined in complex ways, leads tothe desired project outcomes and impacts. Eventhough it is common to distinguish betweenoutcome indicators—which measure the immedi-ate, or short-term, results of projectimplementation—and impact indicators—whichmonitor the longer term or more pervasive resultsof the project—the distinction is sometimesambiguous. For this reason, the approach pre-sented in this note bundles the two together.

Indicators are then developed for both theoverall project objectives and the components:impact indicators relate to the stated objectives ofthe project; output indicators relate to the compo-nents. In the same way that the projectcomponents are closely linked to the overallobjectives of the project, the output and impactindicators should be related. The framework alsoincludes baseline indicators, which monitor aspectsviewed as the context of the project. Table 16presents examples of water indicators to be usedin a project which is structured according to theframework.

The P-S-I-R Approach

Another possible approach to use in the structur-ing of indicators is the Pressure-State-Impact-Response approach. This approach is depicted inFigure 4, and differs a bit from the project basedapproach presented earlier in that it can, withadvantage, be used for activities that focus on arelatively large area (e.g., a whole watershed). Thedifference would be that for a larger area, project-sized objectives and specific components may notbe easily identified. Instead, this frameworkdistinguishes between four different aspects ofenvironmental problems:1) The pressure variable describes the underlying

cause of the problem. It can be an existing

33Conceptual Framework for the Development and Use of Water Indicators

Figure 3. A project based approach

Table 16. Examples of water indicators for a project-based approach

BaselineIndicators

ProjectOutput

Indicators

ProjectOutput

Indicators

ProjectOutput

Indicators

ProjectOutcome/Impact

Indicators

Contextof the

project

The project

and its

componentsObjectives

of theproject

Baseline/output indicatorsBaseline/output indicators Outcome/impact indicatorsOutcome/impact indicators

Water use/demandater use/demandWater recycling potential (%)ater recycling potential (%)Water: annual extraction per capita (mater: annual extraction per capita (m3)Water: annual extraction by sector (%)ater: annual extraction by sector (%)Water: total demand (mater: total demand (m3 3)

Use efficiency (%)Use efficiency (%)Population with access to potaopulation with access to potable water (%)ble water (%)Number of aqueducts (#)Number of aqueducts (#)Water price (US/m )ater price (US/m )

Hydroelectricity generationHydroelectricity generationDams (#, location)Dams (#, location)Hydroelectricity production (mw)Hydroelectricity production (mw)

Kilowatts per hectare inundated (kW)Kilowatts per hectare inundated (kW)

Water emissions/qualityater emissions/qualityN emissions (kg)N emissions (kg)Other emissions (kg)Other emissions (kg)EutrophicationAcidificationToxicity/Heavy metal concentrationoxicity/Heavy metal concentrationColibacilli (m LColibacilli (m L-1)

Biological oxygen demand (mg LBiological oxygen demand (mg L-1)Chemical oxygen demand (mg LChemical oxygen demand (mg L-1)Treatment of used waters (%)reatment of used waters (%)Access to drains (%)Access to drains (%)People affected by diarrheic diseases (#)eople affected by diarrheic diseases (#)

Water availabilityater availabilityWater resater reserves (merves (m3)Rate of recharge (mate of recharge (m3 yr-1)Annual rainfall (mm)Annual rainfall (mm)

Water annual extraction as % of total (%)ater annual extraction as % of total (%)

Population riskPopulation risking inundation (#)Capital risking inundation ($)

Population affected by inundation (#)

Water protectionWatershed land use (ha) Watershed protected area (ha)

(continued)


Table 16. Examples of water indicators for a project based approach (continued)

Population riskPopulation risking inundation (#)Capital risking inundation ($)

Population affected by inundation (#)

Water protectionWatershed land use (ha) Watershed protected area (ha)

Baseline/output indicators Outcome/impact indicators

Figure 4. The Pressure-State-Impact-Response approach

Indicators ofwater use

Indicators ofhydroelectricity

generation

Indicators ofwater demand

Indicators ofwater emissions

Indicators of water availability

Indicators of water quality

Indicators of effects on water

Indicators of population risk

Indicators ofwater

protection

Indicators ofwater

satisfaction

PRESSURE

STATE

IMPACT

RE

SP

ON

SE

35Conceptual Framework for the Development and Use of Water Indicators

problem or it may be the result of a new projector investment.

2) The state variable usually describes somephysical, measurable characteristic of theenvironment that results from the pressure.

3) The impact variable is the same as the impactindicators mentioned above – they monitor thelong-term, or more pervasive, results of aproject or investment.

4) The response variables are those policies,actions or investments that are introduced to

solve the problem. As responses toenvironmental problems they can affect thestate either directly or indirectly, by acting atthe pressures at work.

When indicators are structured according tothe Pressure-State-Impact-Response approach,they become sorted in a somewhat different waythan in the project-based approach. Table 17shows how indicators of use, demand, generation,and emissions are categorized as pressure indica-tors; indicators of availability and quality as state

Table 17. Examples of water indicators for Pressure-State-Impact-Response approach

Detailed informationAggregatedinformation

Indicators of use Annual extraction per capita (m3)Annual extraction by sector (%)

Indicators of demand Total demand (m 3)Use efficiency (%)Recycling potential (%)

Indicators of generationNumber of dams (no)Kilowatts per hectare inundated (kW)Hydroelectricity production (mW)

PRE

SSU

RE

Indicators of emissionsN emissions (kg)Other emissions (kg)

Watervulnerabilityindex

Indicators of availability Reserves (m3)Rate of recharge (m3 yr-1)Annual rainfall (mm)Annual extraction as % of total (%)

STA

TE

Indicators of quality Biological oxygen demand(mg L-1)Chemical oxygen demand (mg L-1)EutrophicationAcidificationColibacilli (m L-1)

Water qualityindex

Indicators of effectsPeople affected by diarrheic diseases (#)Population affected by inundation (#)Toxicity/ Heavy metal concentration

IMPA

CT

Indicators of risk Population risking inundations (no)Capital risking inundations ($)

Climatic riskindex

Indicators of protection Watershed land useWatershed protected area

RE

SPO

NSE

S

Indicators of satisfaction Access to potable water (%)Access to drains (%)Aqueducts (#)Treatment of used waters (%)Water price (US/m3)

Safe waterindex


indicators; indicators of effects and risk as impactindicators; and indicators of protection andsatisfaction as response indicators. Structured inthis way, the indicators in each category can alsobe aggregated into four indices (see Table 17).

Conclusions

This note has presented a conceptual frameworkto the development and use of water indicators, aswell as two different approaches to the structuringof the indicator work—the project-based ap-proach, and the Pressure-State-Impact-Response approach. While there are severalsimilarities to the two, such as the criteria to usein the selection of indicators, the need to identifyproblems as well as establish the appropriateobjectives, and the importance of public consulta-tion, there are clearly times and situations whenone of the frameworks would be preferable to theother.

The first approach presented—the project-basedapproach—is most useful for a well-defined area,or project, where the specification of some project-level objectives and components is feasible. Anexample could be a water purification project—improved access to safe drinking water would bethe project objective, and the installation of watermonitoring stations one of the components. TheP-S-I-R approach, on the other hand, is preferablein a situation that has taken a broader perspective,for example that of a nation, or region. The fourcategories forces the indicator developer and userto consider the various aspects of an environmen-tal problem, which is usually important whenseveral people from various stakeholder groups,are involved in the process.

Another similarity between the two ap-proaches worth pointing out explicitly, is the needfor interpretation of the indicators and for integra-tion in the decision-making process. No indicator—irrespective of the approach used—is the end,but rather the means to an end. That is to say, theintegration of the results from the monitoring intothe decision-making processes, is what reallymakes the difference.

Further ReadingBakkes, J., and J. van Woerden (editors). 1997. The Future of

the Global Environment: A Model-based Analysis SupportingUNEP’s GEO, RIVM-UNEP, Environment Informationand Assessment Technical Report, # 1. Bilthoven, TheNetherlands.

ECLAC. 1992. The Water Resources of Latin America and theCaribbean, LC/R.1120, Water Resources Unit, Divisionof Natural Resources and Energy, ECLAC, Santiago,Chile.

EPA. 1998. The Index of Watershed Indicators,Environment Protection Agency, Washington, D.C.,USA.

Epstein, P. (editor). 1999. Extreme Weather Events, Centerfor Health and the Global Environment, HarvardMedical School, Boston, MA, USA.

Gleick, P. 1998. The World’s Water: The Biennial Report onFreshwater Resources, Island Press, Washington, D.C.,USA.

Hoekstra, A. 1997. The Water Submodel: AQUA, Chapter6, in Rotmans J. and B. de Vries (Editors),Perspectives on Global Change: The TARGETSApproach, Cambridge University Press, CambridgeU.K.

Hoekstra, A., A. Beusen, H. Hilderink, and M. vanAsselt. 1997. Water in Crisis, Chapter 14, in RotmansJ. and B. de Vries (editors), Perspectives on GlobalChange: The TARGETS Approach, CambridgeUniversity Press, Cambridge U.K.

Moldan, B., S. Billharz (editors). 1997. SustainabilityIndicators: Report of the Project on Indicators ofSustainable Development, Wiley and Sons, West Sussex,England.

Revenga, C., S. Murray, J. Abramovitz, and A.Hammond. 1998. Watershed of the World, WorldResources Institute and Worldwatch Institute,Washington, D.C., USA.

Segnestam, L. 1999. Environmental Performance Indicators.A Second Edition Note, The World Bank, Washington,D.C., USA.

United Nations. 1997. Comprehensive Assessment of theFreshwater Resources of the World, WorldMeteorological Organization and StockholmEnvironment Institute, Geneva, Switzerland.

UNEP. 1997. Global Environment Outlook, OxfordUniversity Press, New York, USA.

WCMC. 1998. Freshwater Biodiversity: A PreliminaryGlobal Assessment, by B. Groombridge and M.Jenkins, WCMC Biodiversity Series # 8, WorldConservation Press, Cambridge, U.K.

WHO. 1996. Linkage Methods for Environment and HealthAnalysis: General Guidelines, UNEP-EPA-WHO, WorldHealth Organization, Geneva, Switzerland.

World Bank. 1993. Water Resources Management, A WorldBank Policy Paper, The World Bank, Washington,D.C., USA.

WRI. 1998. World Resources 1998–99: A Guide to the GlobalEnvironment, Oxford University Press, New York, USA.

37

NOTES

1. Collected by CIAT in the rural indicatorsproject and supplemented with data fromother sources.

2. There has been considerable development ofconcepts for criteria and indicators at the forestmanagement level, e.g., for forest certification.A general problem is that these indicatorsreflect good intentions (a perception of what isrelevant) rather than feasibility(measurability).

3. For more information about the ruralsustainability project visit its home page:http://www.ciat.cgiar.org/indicators/index.htm

4. Source: FAO, CCAD and CCAB-AP 1997:“Sustainable forest management in CentralAmerica. Proposal of criteria and indicators atthe forest management unit (FMU).” Food andAgriculture Organization of the UnitedNations (FAO), Central American Commissionon Environment and Development (CCAD),

Central American Council on Forests andProtected Areas (CCAB-AP) (http://www.iucn.org/places/orma/bosques/bosques.html)

5. Zhu, Shushuai , David Tomberlin, and JosephBuongiorno. 1998. “Global forest productsconsumption, production, trade and prices:global forest products model projections to2010.” Paper prepared for the 1999 GlobalForest Products Outlook Study, FAO, Rome(Working paper: GFPOS/WP/01), viii + 37 ppand appendix.

6. The six maps are presented in Annex A.

7. Maps are in color and are best viewed on-screen or in color print. Black and whiteprints will result in the loss of someinformation.

8. For a more detailed discussion about selectioncriteria, see Segnestam (1999), EnvironmentalPerformance Indicators — A Second EditionNote, The World Bank, Washington, D.C., USA.

39

REFERENCES

CCAD (Central American Commission onEnvironment and Development). 1998. Estadodel ambiente y los recursos naturales enCentroamerica 1998. Central AmericanCommission of Environment andDevelopment (p.94). El Salvador.

CIAT. 1999. Accessibility to Markets inHonduras. Cali, Colombia.

CIAT-WB-UNEP. 1999. Forest cover map forCentral America. Cali, Colombia.

COHDEFOR. 1994. Annuario Estadistico Forestal1994. Corporación Hondureña de DesarrolloForestal (COHDEFOR), Departamento dePlanificación: Area Estadisticas Forestales.Tegucigalpa, Honduras.

COHDEFOR. 1997. Annuario estadistico forestal1997. Corporación Hondureña de DesarrolloForestal (COHDEFOR), Departamento dePlanificación: Area Estadisticas Forestales.Tegucigalpa, Honduras.

European Space Research Institute of theEuropean Space Agency (ESA/ESRIN). 1997.Fire Atlas Project: http://shark1.esrin.esa.it/fire_South_America_Txt_Down.html

FAO. 1985. Forest Resources 1980, FAO, Rome.

_____. 1993. FAO Forest Products Yearbook(1993), FAO, Rome.

_____. 1995. Evaluacion de los recursosforestales 1990, FAO, Rome.

FAO, CCAD, and CCAB-AP. 1997. Sustainableforest management in Central America.Proposal of criteria and indicators at theforest management unit (FMU). Food andAgriculture Organization of the UnitedNations (FAO), Central AmericanCommission on Environment andDevelopment (CCAD), Central AmericanCouncil on Forests and Protected Areas(CCAB-AP) (http://www.iucn.org/places/orma/bosques/bosques.html)

INRENARE. 1992. Cobertura Boscosa, 1992.Instituto Nacional de Recursos NaturalesRenovables (INRENARE)

ITTO. 1999. “ITTO Country Profile ofHonduras” (http://www.itto.or.jp/forest_update/v8n2/22.html).

MAG-IMN. 1992. Cobertura de la Tierra, 1992.Original Scale 1:200.000. InsitutoMeteorológico Nacional (IMN-MINAE),Ministerio de Agricultura y Ganadería(MAG).

O.E.A. 1984. Copia Digital Escala 1:500.000Digitised by SIG - PNMC

PAFG. 1996. Cubierta Forestal de la República deGuatemala, 1992. Original Scale 1:500.000.Programa de Acción Forestal para Guatemala(PAFG)

Republica de Honduras. Mapa Forestal. 1995.Corporación Hondureña de DesarrolloForestal (AFE-COHDEFOR). Original Scale1:500,000.


SAR Image (RADARSA). 1998. SAG /DENDRONResource Surveys INC Canada.

Zhu, Shushuai, David Tomberlin, and JosephBuongiorno. 1998. Global forest productsconsumption, production, trade and prices:global forest products model projections to2010. Paper prepared for the 1999 GlobalForest Products Outlook Study, FAO, Rome(Working paper: GFPOS/WP/01), viii + 37 ppand appendix.

World Bank. 1997. Expanding the measure ofwealth—indicators of environmentallysustainable development. EnvironmentallySustainable Development Studies andMonographs No. 17.

World Conservation Monitoring Centre(WCMC). Protected Areas. 1996.

Forest Sector Indicators - World Bankdocuments.worldbank.org › curated › en ›...

Documents

Transcript of Forest Sector Indicators - World Bankdocuments.worldbank.org › curated › en ›...