STATE OF THE WORLD'S FORESTS 2001 PART KEY ISSUES IN THE FOREST SECTOR TODAY
28 STATE OF THE WORLD’S FORESTS 2001 PART II KEY ISSUES IN THE FOREST SECTOR TODAY
PART II KEY ISSUES IN THE FOREST SECTOR TODAY The status of forests. the global forest resources Assessment 2000 D as grown for a broad range began in 1996. In that year, FAO convened an formation on forests at the national and expert consultation in Kotka, Finland, where international levels. Reliable information on the e of the world's leading forest inventory status and trends of forest resources helps give specialists provided technical advice on the scope decision-makers the perspective necessary for and implementation of FRA 2000 as well as on a orienting forestry policies and programmes. Such core set of forest-related definitions to be used in information is useful for monitoring progress the assessment. In 1997. the Fao committee on towards sustainable forest management and for Forestry, FAOs highest-level forest policy forum framing international discussions and agreements for its member countries, and the fourth session on such vital issues as deforestation, biological of the Intergovernmental Panel on Forests(IPF) diversity, desertification, global climate change, approved the consultations finding wood supply and sustainable devele endorsed FAO,s leadership of the assessment. FAO has carried out periodic global forest FRA 2000 was a five-year effort, consisting of a assessments since 1947. at intervals of number of activities a forest assessment based on approximately ten years. This chapter country information; a remote sensing survey of summarizes the methodology and key findings forest cover change at the pan-tropical level on forest area and forest management of FAOs mapping of global forest cover and ecologic most recent and comprehensive forest zones; and the establishment of a forestry assessment, the global Forest Resources information system. FRA 2000 provided basic Assessment 2000(FRA 2000 ). The text also refers assessment information on forest area in 2000 to the two previous assessments: FRA 1990, change in forest area between 1990 and 2000, and which reported on forest cover in 1990 and forest wood volume and biomass. Moreover, in keeping cover change between 1980 and 1990(FAO with recommendations made at Kotka, it 1995a, 1995b), and the interim 1995 assessment, included a number of other parameters to which reported on forest cover in 1995 and provide a more holistic picture of forest resources change from 1990 to 1995(FAO, 1997d). FRA worldwide. The new subjects included, among 2000 forest resource data at country level are other things, forest area under protection status, provided in Annex 2. Detailed country profiles, trees outside forests, forest fires, non-wood forest including baseline data, and material on all products, timber removals and information on parameters covered by the assessment, are available on the FAO Forestry Department Web Great effort was made to ensure that the Fra site. The full report of FRA 2000 is due to be process was both participatory and transparent. published by the end of 2001 National forestry agencies from nearly every THE ASSESSMENT PROCESS The global forest resources assessment 2000 I The FAO Forestry Department Web site is the most up-to-date source of information on all aspects of the assessment and was a joint endeavour carried out by FAO in containsinformationonallcountries(www.fao.org/forestry/fo/ fra/index jsp). The results of the assessment of temperate and cooperation with its member countries and many boreal forests in industrialized countries are also available in other partners. detailed planning for FRA 2000 UN-ECE/FAO(2000b)
PART II KEY ISSUES IN THE FOREST SECTOR TODAY 29 Demand has grown for a broad range of information on forests at the national and international levels. Reliable information on the status and trends of forest resources helps give decision-makers the perspective necessary for orienting forestry policies and programmes. Such information is useful for monitoring progress towards sustainable forest management and for framing international discussions and agreements on such vital issues as deforestation, biological diversity, desertification, global climate change, wood supply and sustainable development. FAO has carried out periodic global forest assessments since 1947, at intervals of approximately ten years. This chapter summarizes the methodology and key findings on forest area and forest management of FAO’s most recent and comprehensive forest assessment, the Global Forest Resources Assessment 2000 (FRA 2000). The text also refers to the two previous assessments: FRA 1990, which reported on forest cover in 1990 and forest cover change between 1980 and 1990 (FAO, 1995a, 1995b), and the interim 1995 assessment, which reported on forest cover in 1995 and change from 1990 to 1995 (FAO, 1997d). FRA 2000 forest resource data at country level are provided in Annex 2. Detailed country profiles, including baseline data, and material on all parameters covered by the assessment, are available on the FAO Forestry Department Web site.1 The full report of FRA 2000 is due to be published by the end of 2001. THE ASSESSMENT PROCESS The Global Forest Resources Assessment 2000 was a joint endeavour carried out by FAO in cooperation with its member countries and many other partners. Detailed planning for FRA 2000 The status of forests: the Global Forest Resources Assessment 2000 began in 1996. In that year, FAO convened an expert consultation in Kotka, Finland, where some of the world’s leading forest inventory specialists provided technical advice on the scope and implementation of FRA 2000 as well as on a core set of forest-related definitions to be used in the assessment. In 1997, the FAO Committee on Forestry, FAO’s highest-level forest policy forum for its member countries, and the fourth session of the Intergovernmental Panel on Forests (IPF) approved the consultation’s findings and endorsed FAO’s leadership of the assessment. FRA 2000 was a five-year effort, consisting of a number of activities: a forest assessment based on country information; a remote sensing survey of forest cover change at the pan-tropical level; mapping of global forest cover and ecological zones; and the establishment of a forestry information system. FRA 2000 provided basic assessment information on forest area in 2000, change in forest area between 1990 and 2000, and wood volume and biomass. Moreover, in keeping with recommendations made at Kotka, it included a number of other parameters to provide a more holistic picture of forest resources worldwide. The new subjects included, among other things, forest area under protection status, trees outside forests, forest fires, non-wood forest products, timber removals and information on forest management. Great effort was made to ensure that the FRA process was both participatory and transparent. National forestry agencies from nearly every 1 The FAO Forestry Department Web site is the most up-to-date source of information on all aspects of the assessment and contains information on all countries (www.fao.org/forestry/fo/ fra/index.jsp). The results of the assessment of temperate and boreal forests in industrialized countries are also available in UN-ECE/FAO (2000b)
STATE OF THE WORLD's FORESTs 2001 country of the world, a large number of research and FAO staff and consultants visited countries centres and academic institutions and several and held workshops with concerned country international, regional and non-governmental representatives. FAO relied mainly on statistics organizations collaborated in the implementation from primary sources, rather than on quoted or secondary sources. In the countries where no data, FAO formally invited countries to review applicable national forest inventories existed, it the preliminary results of the assessment. was necessary to piece together information from Countries were given the opportunity to submit various partial inventories or to use secondary comments and supporting technical material that sources and reconnaissance surveys. In contrast, could improve the results compiled by FAO UN-ECE collected information on industrialized ckground information and analyses used in the countries from questionnaires filled out by national forest resource calculations have been made correspondents, who used mainly national forest available on the FAO Forestry Department Web inventory data All data provided to FAO and site(www.fao.org/forestry/fo/country/indexjsp),Un-eCewerecheckedandvalidatedwiththe naking it possible to trace the final results back national correspondents in an intense dialog to the original source data. over nearly two years. everal major challenges had to be met in order Forest assessment based on country to assemble country information and then information integrate it to form a global picture of the status of The central undertaking of FRa 2000 was the forest area in 2000 and change in forest area forest assessment based on country information. between 1990 and 2000. After FAO and its National-level data on forest resources were partners made a major effort to amass all relevant collected through an exhaustive survey of inventory and related information, the information inventory reports and other information from base for many countries was found to be limited. countries. National forestry experts and other For example, over half of the developing countries artners from around the world were involved in had only one forest inventory, and more than one this effort. A major partner was the UN fourth of them had never carried out an inventory Economic Commission for Europe(UN-ECE (see Table 1). Most of the country data used for UN-ECE coordinated the assessment of temperate FRA 2000 spanned a period of about ten years and tropical forests in the industrialized countries (although a few inventories were even older) and countries in transition: australia, the Only a handful of countries maintain continuous Commonwealth of Independent States( CIS), ational forest inventories with comparable time Europe, Japan, New Zealand and north series. As a result, it was difficult to calculate America. fao coordinated the assessment of precise estimates of forest change at both the developing countries, in which almost all tropical national and global levels. Projecting data and subtropical forests and some temperate forwards and backwards to the reference years forests are located, and was also responsible for 2000 and 1990 was a critical and difficult aspect of the integration of all the information into the the assessment. In the absence of comparable harmonized synthesis constituting the global multiple-date inventories, a trend line had to be derived for many countries by using a variation of To collect data for the developing countries, the"convergence of evidence"method, in which FAO formally requested country representatives to countries' survey results were complemented with supply the most recent forest inventory reports other information such as inventory statistics, economic information and policy studies. Finally, forest vegetation types and terminology varied The counties considered to be in the European region for the widely among countries, compounding the purposes of the assessment coordinated by the UN-ECE include those listed under Europe in Annex 2 except members of the problem of aggregating national data into CIS, plus Cyprus, Israel and Turkey
30 STATE OF THE WORLD’S FORESTS 2001 and FAO staff and consultants visited countries and held workshops with concerned country representatives. FAO relied mainly on statistics from primary sources, rather than on quoted or secondary sources. In the countries where no applicable national forest inventories existed, it was necessary to piece together information from various partial inventories or to use secondary sources and reconnaissance surveys. In contrast, UN-ECE collected information on industrialized countries from questionnaires filled out by national correspondents, who used mainly national forest inventory data. All data provided to FAO and UN-ECE were checked and validated with the national correspondents in an intense dialogue over nearly two years. Several major challenges had to be met in order to assemble country information and then integrate it to form a global picture of the status of forest area in 2000 and change in forest area between 1990 and 2000. After FAO and its partners made a major effort to amass all relevant inventory and related information, the information base for many countries was found to be limited. For example, over half of the developing countries had only one forest inventory, and more than onefourth of them had never carried out an inventory (see Table 1). Most of the country data used for FRA 2000 spanned a period of about ten years (although a few inventories were even older). Only a handful of countries maintain continuous national forest inventories with comparable time series. As a result, it was difficult to calculate precise estimates of forest change at both the national and global levels. Projecting data forwards and backwards to the reference years 2000 and 1990 was a critical and difficult aspect of the assessment. In the absence of comparable multiple-date inventories, a trend line had to be derived for many countries by using a variation of the “convergence of evidence” method, in which countries’ survey results were complemented with other information such as inventory statistics, economic information and policy studies. Finally, forest vegetation types and terminology varied widely among countries, compounding the problem of aggregating national data into harmonized global estimates. country of the world, a large number of research centres and academic institutions and several international, regional and non-governmental organizations collaborated in the implementation of FRA 2000. Before publishing the FRA 2000 data, FAO formally invited countries to review the preliminary results of the assessment. Countries were given the opportunity to submit comments and supporting technical material that could improve the results compiled by FAO. Background information and analyses used in the forest resource calculations have been made available on the FAO Forestry Department Web site (www.fao.org/forestry/fo/country/index.jsp), making it possible to trace the final results back to the original source data. Forest assessment based on country information The central undertaking of FRA 2000 was the forest assessment based on country information. National-level data on forest resources were collected through an exhaustive survey of inventory reports and other information from countries. National forestry experts and other partners from around the world were involved in this effort. A major partner was the UN Economic Commission for Europe (UN-ECE). UN-ECE coordinated the assessment of temperate and tropical forests in the industrialized countries and countries in transition: Australia, the Commonwealth of Independent States (CIS), Europe,2 Japan, New Zealand and North America. FAO coordinated the assessment of developing countries, in which almost all tropical and subtropical forests and some temperate forests are located, and was also responsible for the integration of all the information into the harmonized synthesis constituting the global assessment. To collect data for the developing countries, FAO formally requested country representatives to supply the most recent forest inventory reports, 2 The countries considered to be in the European region for the purposes of the assessment coordinated by the UN-ECE include all those listed under Europe in Annex 2 except members of the CIS, plus Cyprus, Israel and Turkey
PART II KEY ISSUES IN THE FOREST SECTOR TODAY To make the highly variable country ssessment to use it as the minimum cano information useful for global reporting, FAo cover to describe forests in industrialized developed a set of protocols and standards for countries as well. its harmonization. All country information had To make the comparison between forest area first to be classified according to a common set in 1990 and 2000 possible the 1990 national of terms and definitions. (See Annex 1 for some forest area figures derived by the interim 1995 definitions used. )This was a difficult task owing assessment were revised, using the same to the sheer magnitude and variability of the definitions, methodologies and new inventory information produced by countries and the wide data used for calculating the 2000 figures. This range of forest formations, ecological conditions established a new baseline for forest cover in and cover types that exist worldwide. For 1990. It should be noted that updating national example, FRA 2000 assembled more than 650 forest area figures from a previous forest definitions of forest from 132 developing resources assessment by incorporating new data countries. Reducing this information into a is a standard feature of alll fao forest (i.e. closed forest, open forest, and other wooded land)was a major task. For the definition of Pan-tropical remote sensing survey forest, FAO adopted the threshold of 10 percent An independent remote sensing survey was crown cover. recommended in the landmark conducted for FRA 2000 to complement the tudy on worldwide vegetation classifications survey based on country information. Controlled carried out by the United Nations Educational, mpling of tropical forests combined with a Scientific and Cultural Organization(UNESCO, uniform data source -satellite imagery 1973), this threshold was used for developing provided a comparable set of statistics for 1980- countries in the Forest Resources Assessments of 1990 and 1990-2000, making possible a direct 1980 and 1990. but fra 2000 was the first comparison of forest area change over these two TABLET Forest inventories in developing countries as with a national with no forest vith a partial forest inventory. After 1990 Before 1990 Asia and oceania 3 Latin America Totals he term "repeated"is used to refer to continuous monitoring or comparable inventories carried out at fixed intervals. The term single shot"refers to countries that have carried out either one or multiple inventories that are not comparable with one another. The date(after or before 1990) refers to the most recent inver
PART II KEY ISSUES IN THE FOREST SECTOR TODAY 31 To make the highly variable country information useful for global reporting, FAO developed a set of protocols and standards for its harmonization. All country information had first to be classified according to a common set of terms and definitions. (See Annex 1 for some definitions used.) This was a difficult task owing to the sheer magnitude and variability of the information produced by countries and the wide range of forest formations, ecological conditions and cover types that exist worldwide. For example, FRA 2000 assembled more than 650 definitions of forest from 132 developing countries. Reducing this information into a highly compressed set of global forest classes (i.e. closed forest, open forest, and other wooded land) was a major task. For the definition of forest, FAO adopted the threshold of 10 percent crown cover. Recommended in the landmark study on worldwide vegetation classifications carried out by the United Nations Educational, Scientific and Cultural Organization (UNESCO, 1973), this threshold was used for developing countries in the Forest Resources Assessments of 1980 and 1990, but FRA 2000 was the first assessment to use it as the minimum canopy cover to describe forests in industrialized countries as well. To make the comparison between forest area in 1990 and 2000 possible, the 1990 national forest area figures derived by the interim 1995 assessment were revised, using the same definitions, methodologies and new inventory data used for calculating the 2000 figures. This established a new baseline for forest cover in 1990. It should be noted that updating national forest area figures from a previous forest resources assessment by incorporating new data is a standard feature of all FAO forest assessments. Pan-tropical remote sensing survey An independent remote sensing survey was conducted for FRA 2000 to complement the survey based on country information. Controlled sampling of tropical forests combined with a uniform data source – satellite imagery – provided a comparable set of statistics for 1980- 1990 and 1990-2000, making possible a direct comparison of forest area change over these two Regions Number of Countries/areas Countries/areas Countries/areas with a national countries/areas with no forest with a partial forest inventory: inventory forest inventory Repeated1 Single shot2 After 1990 Before 1990 Africa 56 14 15 7 12 10 Near East 13 11 0 2 0 0 Asia and Oceania 22 2 2 3 13 3 Latin America 21 1 4 15 16 11 Caribbean 24 13 0 4 6 10 Totals 136 39 21 32 47 34 1 The term “repeated” is used to refer to continuous monitoring or comparable inventories carried out at fixed intervals. 2 The term “single shot” refers to countries that have carried out either one or multiple inventories that are not comparable with one another. The date (after or before 1990) refers to the most recent inventory. TABLE 1 Forest inventories in developing countries
STATE OF THE WORLD's FORESTs 2001 sessment periods. The survey relied on FaoForestryDepartmentWebsite(www.fao.org/ statistical sampling (10 percent) of 87 percent of forestry/fo/fra/index sp) the world's tropical forests through 117 sample The accuracy of the forest map is estimated at units to produce estimates of the status and about 80 percent for all forest classes. Accuracy change of tropical forest at regional, ecological for closed forests is somewhat higher and and pan-tropical levels (but not at the national accuracy for open/ fragmented forests level). The principal output of the remote somewhat lower Other wooded lands has the sensing survey was an area change matrix, which lowest accuracy of the three woody vegetatio illustrates and quantifies how the forest and other classes represented land use classes changed between 1980 and 2000. The global ecological zoning map provided a The forest and land cover classification scheme means of differentiating forests globally by used by the remote sensing survey was linked ecological zone. While most countries have closely to the Fra forest classes used for the appropriate means of compiling national country-based survey and for the low-resolution information on forests according to ecological global forest map, so that data from these three units, it was not possible to aggregate this information at the global level prior to the development of the ecological zoning map FRA 2000 global maps a major reason for this was the absence of an The production of global maps was a significant internationally accepted global standard and new undertaking for FRA 2000. For the first classification system that is geometrically correct ime, a global forest map now exists that shows and registered to a map base. In the past, few the location and distribution of forests according applications have required analysis and reporting to FRA classifications. FRA 2000 also produced according to ecological zone at the global scale, global maps on ecological zones and protected and until now, useful global maps have been areas. The FRA 2000 maps are useful visual aids slow to emerge. However, because certain for understanding the location and extent of the environmental functions have intemational major forest areas of the world. Each map is dimensions, global applications of ecological generated using computerized Geographic zoning are expected to become increasingly Information System(GIS) technology, which makes it possible to combine the maps with FAO identified the Koppen system other spatial and statistical data. modified by Trewartha in 1968, as the most The forest map has been printed at a scale of appropriate ecological zoning scheme for the Fra 1: 40000 000 and enlargements up to 1: 10000 000 2000 ecological zoning map. Slightly modifying are possible. a poster version of the forest map this scheme, FAO identified 20 global ecologica accompanies this publication (reproduced in zones, ranging from evergreen tropical rain forest Figure 2). Digital versions are available on the to boreal tundra (see box 12 ). These were then mapped. The worldwide forest cover according to ecological zone was determined by overlaying 3 The Kotka expert consultation (referred to as Kotka m) advised the Fra 2000 global forest cover map on the FAO to consider conducting the remote sensing survey at the global ecological zoning map in the GIS,and however, limited the work to the tropics, except for some pilot then extracting the statistics. FAO developed a network of cooperators who were instrumental in the development of the maps. Major collaborators involved in the effort induded eroS Data Center of the united s a small number of dassification schemes have been States, the World Conservation Monitoring Centre, based in the eloped for use at the global level, including Bailey. United Kingdom: the Institute for Applied Research and Holdridge, Koppen and Thormwaite, but none of these is Analysis, in Austria, the Laboratory of Terrestrial Ecology of available digitally or is registered to a geometrically correct map France: the Canadian Center for Remote Sensing: the United base. Koppen's classification scheme has been the most tates Forest Service: and the Australian Bureau of Rural Sciences. widespread and longest used
32 STATE OF THE WORLD’S FORESTS 2001 assessment periods. The survey relied on statistical sampling (10 percent) of 87 percent of the world’s tropical forests through 117 sample units to produce estimates of the status and change of tropical forest at regional, ecological and pan-tropical levels (but not at the national level).3 The principal output of the remote sensing survey was an area change matrix, which illustrates and quantifies how the forest and other land use classes changed between 1980 and 2000. The forest and land cover classification scheme used by the remote sensing survey was linked closely to the FRA forest classes used for the country-based survey and for the low-resolution global forest map, so that data from these three sources are complementary. FRA 2000 global maps The production of global maps was a significant new undertaking for FRA 2000.4 For the first time, a global forest map now exists that shows the location and distribution of forests according to FRA classifications. FRA 2000 also produced global maps on ecological zones and protected areas. The FRA 2000 maps are useful visual aids for understanding the location and extent of the major forest areas of the world. Each map is generated using computerized Geographic Information System (GIS) technology, which makes it possible to combine the maps with other spatial and statistical data. The forest map has been printed at a scale of 1:40 000 000 and enlargements up to 1:10 000 000 are possible. A poster version of the forest map accompanies this publication (reproduced in Figure 2). Digital versions are available on the FAO Forestry Department Web site (www.fao.org/ forestry/fo/fra/index.jsp). The accuracy of the forest map is estimated at about 80 percent for all forest classes. Accuracy for closed forests is somewhat higher and accuracy for open/fragmented forests is somewhat lower. Other wooded lands has the lowest accuracy of the three woody vegetation classes represented. The global ecological zoning map provided a means of differentiating forests globally by ecological zone. While most countries have appropriate means of compiling national information on forests according to ecological units, it was not possible to aggregate this information at the global level prior to the development of the ecological zoning map. A major reason for this was the absence of an internationally accepted global standard and classification system that is geometrically correct and registered to a map base.5 In the past, few applications have required analysis and reporting according to ecological zone at the global scale, and until now, useful global maps have been slow to emerge. However, because certain environmental functions have international dimensions, global applications of ecological zoning are expected to become increasingly important. FAO identified the Köppen system, as modified by Trewartha in 1968, as the most appropriate ecological zoning scheme for the FRA 2000 ecological zoning map. Slightly modifying this scheme, FAO identified 20 global ecological zones, ranging from evergreen tropical rain forest to boreal tundra (see Box 12). These were then mapped. The worldwide forest cover according to ecological zone was determined by overlaying the FRA 2000 global forest cover map on the global ecological zoning map in the GIS, and then extracting the statistics. 3 The Kotka expert consultation (referred to as Kotka III) advised FAO to consider conducting the remote sensing survey at the global level with about 350 sample units. Financial restrictions, however, limited the work to the tropics, except for some pilot activities. 4 FAO developed a network of cooperators who were instrumental in the development of the maps. Major collaborators involved in the effort included EROS Data Center of the United States; the World Conservation Monitoring Centre, based in the United Kingdom; the Institute for Applied Research and Analysis, in Austria; the Laboratory of Terrestrial Ecology of France; the Canadian Center for Remote Sensing; the United States Forest Service; and the Australian Bureau of Rural Sciences. 5 A small number of classification schemes have been developed for use at the global level, including Bailey, Holdridge, Köppen and Thornwaite, but none of these is available digitally or is registered to a geometrically correct map base. Köppen’s classification scheme has been the most widespread and longest used