The United Nations Food and Agriculture Organization regularly reports on the world's forest resources through the Forest Resources Assessment Programme. The first assessment was initiated in 1946. FAO has since then conducted Global Forest Resources Assessments every 5-10 years.
The most recent Global Forest Resources Assessment and hitherto the most comprehensive is the Global Forest Resources Assessment 2000 (FRA 2000) which describes the state of the worlds forest by the year 2000. The web version of FRA 2000 was made available on the 01.10.2001 via the FAO Forestry website (http://www.fao.org/forestry/fo/fra/main/index.jsp).
Of all parameters reported in the global forest resources assessments, forest area and in particular forest area change rate get the most attention. There has been considerable interest in comparing the current estimates with the previous assessment (FRA 1990). However a number of factors make this comparison difficult. This paper explains these differences and provides comparisons of the two latest assessments, related to forest area and area change.
The FRA 2000 methodology is described in FAO 2001a and UNECE/FAO 2000 as well as the FRA working papers listed at the end of the paper. The FRA 1990 methodology is described in FAO 1993, 1995a and 1995b. The following highlights some points important to this paper.
_ While technical and scientific improvements have greatly increased the potential to improve the information base, many countries still lack the basic training, institutional and financial resources to conduct periodic assessments. The average reference year of the underlying national source data provided for the FRA 2000 area and forest area change estimates is 1994 with considerable older data for some developing countries. The approximate average reference year for country data from developing countries made available for FRA1990 was 1980.
_ FRA 1990 area and forest area change estimates for the developing countries were generated by use of a mathematical deforestation model (or forest area adjustment model) partially driven by population density and growth (FAO 1993, FAO 1995b).
_ Use of the deforestation model was discontinued in FRA 2000 after the acknowledgement that there were relatively weak correlations at national level between forest area change rates and demographic parameters (FAO 2001a). FRA 2000 area and forest area change estimates do not build on earlier assessments and they are not generated by models. The FRA 2000 figures for forest area and forest area change are estimates for each individual country based on the data from the submitted country reports. Along with the core information on the state and changes in forest areas, FRA 2000 contains a number of special studies on topics like ecological zones, forest fires, plantations and non-wood forest products.
_ The FRA 2000 pan tropical remote sensing survey (RSS) was performed independently of the compilation of country data and was used to complement and validate the data of the tropical regions. A representative sample of tropical forests was covered over the period 1980-2000 and the FRA 2000 RSS built on the methodology and experiences from FRA 1990.
_ FRA 2000 main report (FAO Forestry Paper 140) points at a lower rate of net forest area loss world wide in the 1990's than in the 1980's, owing mainly to a higher rate of natural expansion of forest area during the 1990's. It is also mentioned that the worldwide loss of natural forest has continued at roughly comparable high levels over the past 20 years.
The present working paper is compiled to verify the conclusions concerning forest area and forest area change of FRA 2000 main report.
The present working paper compares and discusses the figures for forest area and annual net forest area change for the periods 1980-1990 and 1990-2000 as reported by FRA 1990 and FRA 2000 respectively. The figures compared are based on the submitted country reports and taken directly from FAO Forestry Paper 124 and 140 respectively. The findings of FRA 2000 RSS have been used to compare the overall trends and for regional adjustment of the forest area changes reported for Africa described in section 4.2. Comparisons between FRA 1990 and FRA 2000 figures is complicated as the assessments use different definitions of forests (applies especially for the industrialized countries), different reference years and varying quality of source data and, in the case of developing countries, a different approach for generating the FRA 1990 figures for forest area and forest area change.
Two comparisons are made:
1. The area of forest in 1990. This area was estimated both by FRA 1990 and FRA 2000. The working paper determines the level of difference between the two estimates and identifies the main causes and contributors.
2. The figures for change in forest area 1980-1990 and 1990-2000 indicate a lower net change for FRA 2000 than for FRA 1990. The working paper analyses the validity of a comparison of the figures for forest area change.
The working paper is structured as follows.
Chapter 2 compares Area90FRA1990 with Area90FRA2000. The main contributors to the difference between the figures are identified and the findings at subregional level are briefly discussed.
Chapter 3 undertakes a similar process of identification and discussion of the main contributors to the observed difference between AreaChangeFRA1990 and AreaChangeFRA2000
Chapter 4 analyses
_ Effect of changed definitions on Forest Area and Forest Area Change
_ Effect of Adjustments for plantations success rates on Forest Area Change and adjustment for forest area change figures for Africa based on the findings of FRA 2000 RSS.
_ Effect of Improved Data from FRA1990 to FRA 2000 on forest area change estimates
Chapter 5 concludes the study by summing up the main findings, pointing out the acknowledged weaknesses and pointing to areas for future study.