FRA 2000 prepared and presented a background paper3 on possible information sets that could be used for the forest-change prediction algorithm. The experts were asked to review the paper and evaluate the proposed options for the forest change estimation procedure in anticipation of the workshop. In discussing these at the meeting, the participants were asked to consider these and other approaches based on their experience and expertise. .
Given the scope and complexity of FRA 2000, various approaches to data capture are needed to cover all aspects of the assessment. Three basic and complementary information generation mechanisms, or modules, available to FRA 2000 are a) an assessment based on existing information, b) an assessment based on newly generated information through a global remote sensing survey and c) special studies.
The collection of existing information is a continuous process. For every new assessment the data sources used in the previous assessments are consulted and new sources identified through contacts in the countries. When new sources are identified, they are compared with the older information to determine which provides the best and most reliable baseline4.
Comparability5 between two (or more) information sources is also evaluated, verifying their utility to serve as representative surveys in a continuous time-series. Information sources for the FAO assessments typically include:
Frequently, many institutions contribute all the information needed for an assessment. These include forestry agencies (particularly their inventory, statistics or mensuration units), environmental and natural resources agencies, resource survey units and mapping and remote sensing institutions. Other organizations, such as regional development agencies, may also provide relevant information. With this in mind, data compilers need to check with a range of national agencies and organizations to ascertain the existence of new information.
Once the new data are transmitted to FAO they are reviewed and recorded into a bibliographic reference system and archived in the FRA Programme’s library.
Many countries that have forested land are interested in the status and trends of forest cover, growth, mortality and removals. Some countries have in place systems for monitoring their forests on a periodic or occasional basis. Other countries have conducted studies of subsets of their forests; for example, studies of opportunities for potential commercial operations. Many developing countries have undertaken complete or partial forest inventories sponsored by a variety of developmental organizations including FAO, the International Bank of Reconstruction and Development (World Bank) and various international donor agencies.
The result is a huge set of information regarding forest status and trends at the national and sub-national level. If all of this information was current, reliable, aggregated in a single location and collected to a standard format, it would possible to readily incorporate it into a report of forest change for FRA 2000. In reality, such information is scattered in the various countries, development agencies and donor programs across the world, and is difficult to access. Nonetheless, FRA 2000 professional staff have been working for years to make and keep current a meta-database describing what data exist, by country, throughout the tropical world, as well as archiving the actual data in the FORIS (Forest Resources Information System) database (see section 6.4). Regional specialists travel to the countries in their region, gleaning available information and assembling it for potential incorporation into FRA 2000 and future studies
The strength of using data reported by countries is that results tend to be of higher precision than results from studies that only use regional data, such as the FRA 1990 remote sensing survey (see FRA 1990).
The primary weakness in the aggregation of country level inventory data is that there is tremendous inconsistency and incompleteness, both within countries over time as well as between countries. Assessments are conducted with different objectives and populations of interest, using different definitions, standards, sample designs and field methods from those used by FAO or by neighbouring countries. The results can be difficult to interpret and harmonize for comparison across countries, as FRA must do. Additionally, many countries, particularly in Africa, simply lack the resources to conduct forest assessments, so data are lacking.
FRA 1990 used a statistical model to generate estimates of forest change for countries. The strength of a modelling approach such as this is that it provides a platform to generate estimates for all countries regardless of availability of country-level data. Data gaps would simply be modelled by more generalized functions. Generation of models is a useful exercise in that it forces the study of cause and effect relationships and can lead to insights regarding true mechanisms or drivers of forest cover change, which in turn could lead to better investment decisions by national governments seeking to manage their forests.
The results of existing models are probably reasonable at the global level. However, the greatest weakness of the modelling approach continues to be the inadequacy of existing models when applied to specific countries. Deforestation is such a complex process involving physical, climatological, political and socio-economic forces that are themselves very complex, that no simple generalized models of forest change have been developed that produce satisfactory results for all countries. Current models are oversimplified and yield similar predictions of forest cover change rates for countries that are known to be very different. More complex models are yet to be developed and tested.
The FRA 1980 estimation method relied heavily on a review of existing sources, complemented by the opinions of experts experienced in the countries. These experts evaluated the reliability of the existing sources and contributed additional knowledge and information on the situation for each country in order to identify factors and processes involved in forest change. National forest cover state and change were then subjectively estimated through expert opinion and review of the information.
The strength of the FRA 1980 method lies in the large range and amount of supplementary information the experts provided through their in-depth knowledge of the countries’ forests, arriving at relatively precise statements. The disadvantage of using expert opinions was the implied bias (deviation from true values) that always accompanies subjective estimates. Nevertheless, this approach represented the state of the art for global assessments at the time, and the range of qualitative information produced by the assessment is unmatched today, though much is now outdated.
FRA 2000 recognizes the importance of developing and improving methods to collect, analyse and process data from a variety of sources, including remote sensing information, to assess the change in forest cover world-wide at the national, regional and global levels. The experts were urged to provide concrete recommendations and advice about methods to improve the FRA programme.
In regard to the harmonization of FAO and country standards and definitions for such things as “forests”, “forest degradation” and “deforestation”, it was indicated that both local and FAO official terms and definitions will be entered into the database, which may aid in eventual harmonization.
The key end product of the FRA 2000 process for each country, its country profile web page, contains information divided into 11 subjects. It is currently under construction and available only within FAO. Data and definitions included in the web site derive from different sources, but mainly from the countries. After the information is entered it is sent to countries for feedback and validation
2 Section number 2 was taken from: Päivinen. R. and Gillespie, A. 2000. Estimating Global Forest Change:1980-1990-2000. A background document prepared for an international panel of experts convened to review methods to be used in completing the FAO Global Forest Resource Assessment (FRA 2000), Summary Section iii. FAO, Rome.
3 See background document, Appendix #2
4 Baseline refers to the inventory or assessment used for estimating forest area according to FAO standard classification and reference year. The single most appropriate assessment is always used as the baseline. Appropriateness is a function of reliability, currency and compatibility with the FAO assessment objectives.
5 Comparability between assessments requires that the techniques and protocols employed in the various surveys are sufficient to produce findings that can be compared. Few countries today, even in industrialized regions, are generating comparable multi-date information as techniques and assessment objectives frequently change from survey to survey.
