During the Expert Consultation Meeting additional presentation were given by participants and FRA staff.
This project aims to map the extension of deforestation in Brazilian Amazonia, using 1997as a base date. It mapped the additional deforestation and the regenerated areas for 1998 and characterized the fires in deforested areas. The project will built a data bank with digital maps for Amazonia.
The system looks at about 190 TM scenes out of 240 per year. Most deforestation is concentrated in about 40 scenes. They produce a preliminary estimate based on the sample of 40, then complete an analysis of about 150 more (the remaining 50 or so scenes are typically occluded by clouds). The difference between the sample estimate and the final estimate has been about 1.5% in the past two years.
The reported information is total hectares cleared each year with deforestation measured in terms of land cleared, based on cover (loss of primary forest). Brazil’s approach is cumulative and does not recognise areas that ultimately stay as forest (e.g. regrow into secondary forest).
Brazil does have estimates of the rates at which the cleared land goes into agriculture, pasture and forest, so they should be able to respond to FRA requests. The FRA definition is based on hectares taken out of forest land use. In the FRA report, it was recommended that attention be paid to possible positive changes and their impact on the figures to be reported in FRA 2000.
A short-time series of cover maps and figures from the 1940s to 1990s shows a general decline in cover from 1940 to 1990, then a sharp increase since 1990. Estimates were based on varying degrees of information: sampling, remote sensing and personal experience based on a two-week visit.
It was pointed out that the FRA global forest cover map from the EROS Data Center not very reliable at the country level in Costa Rica: it misses significant forest in the dry forest zone and the mountain forest zone; it adds significant forest in the urban central valley zone and it includes agricultural lands (e.g. banana plantations).
It was suggested that the web page should label the maps that are preliminary or under revision.
FRA 2000 will present new global maps for forest cover, ecological zones and protected areas. Spatial information about the world’s forests is an important prerequisite for global environmental change research, from studies on biodiversity and effective conservation to modeling sustainability of forest ecosystems.
The forest cover map presents the first comprehensive worldwide view of forests using a consistent methodology and standard data. The map has a resolution of 1 km and is based on 1992-1993 and 1995-1996 AVHRR data. Four broad land cover categories are presented following FAO’s standard classification: closed forest, open/fragmented forest, other wooded land and other land. The primary use of the map is to show the current extent of forests at a global and regional level. Other potential uses include forest change assessment and modelling, and to supplement regions lacking recent, reliable forest inventory data. Due to the coarse source data, the map cannot be used to obtain reliable country forest cover statistics. The final map has been drafted through validation with information/maps based on higher resolution data, i.e. Landsat TM and SPOT. An accuracy assessment will be carried out later this year to indicate the precision of the map.
The map was developed at the EROS Data Center (EDC) of the United States of America. The World Conservation Monitoring Center (WCMC) in the United Kingdom collaborated in the map validation.
Along with the core information on the state and changes in forests, FRA 2000 is reporting on various ecological aspects of forests. Forest resources information will be reported by ecological zone, which will contribute to understanding the implications of forest change on (ecosystem) biological diversity and carbon-cycling processes.
The 2000 global ecological zone map has been developed, building upon the FRA 1990 experience for the tropics and extending the coverage to include the temperate and boreal forests. A globally consistent classification has been adopted, based on the Köppen-Trewartha climate system in combination with natural vegetation characteristics. A total of 19 global ecological zones have been defined and mapped, ranging from the evergreen tropical rainforest zone to the boreal tundra woodland zone. A main principle of delineating the global ecological zones involved the aggregation or matching of available regional ecological or potential vegetation maps into the global framework.
The main uses of the map include: a) reporting purposes, to provide baseline forest statistics by ecological zone; b) analysis and modelling purposes, i.e. assessment of forest-based biological diversity, both at ecosystem and species level, and biomass modelling.
FAO’s main partners in the development of the Ecological Zone map were the WCMC, EDC and the Laboratoire d’Ecologie Terrestre (LET) of France. In addition, regional experts and scientists have been instrumental in the production of the map by providing their expertise, advice and input data. A key event was the Cambridge expert consultation (July 1999), attended by some 20 regional ecozone experts, where the proposed classification framework was presented and adopted.
This map shows the location and extent of the world’s protected areas as of January 2000. Along with the map data, information is provided for each country on the name, area and category (IUCN categories I – VI, International Conventions) of each protected area. A main use of the map is to show the conservation status of the world’s forests.
The map was produced by WCMC, based on data/information provided by the countries.
The Forest Resources Information System (FORIS) provides the world with baseline information on forest resources. FORIS was developed for the production and presentation of the Global Forest Resources Assessment 2000 (FRA 2000), but with the perspective to continue to improve it beyond this event. FORIS includes both existing source information and derived information.
The following points have guided the system development work:
Presently, ten thematic subjects have been identified. These are: country description, forest cover, plantations, volume/biomass, protected areas/conservation, forest change, ownership, non-wood goods and services, fires and wood supply. For each of these subjects a number of subject classes are identified. The subject classes essentially represent the final output of FRA 2000 (or any following presentation). For example, the forest cover subject includes the subject classes composition, disturbance and texts.
The process for incorporation of data into FORIS is shown in the next figure:
Figure 1. Information production process
The development of an Internet application in FORIS has not started. A mockup design of the contents of country pages has been done and will be used as a template. People to take part in the work have been identified. There will be close links to the Forestry web site.
The main issues related to a World Forest Survey (WFS) were presented:
The major forestry information requirements of a WFS, in a very general sense, are to provide supply and demand functions of products and services that are relevant (and cost efficient) for decisions and have known precision and accuracy.
There are several possible components of a world forest survey design:
A world forest survey requires well-defined customers, a very flexible design, access to all countries and broad institutional collaboration. It is a huge task. FAO, with the experience gained through FRA 2000, could possibly serve as secretariat for this process.
Current activities and approaches:
There was a general consensus that a world forest survey is a worthy goal. One way to facilitate success would be to build as much as possible on existing in situ inventory and monitoring systems at the country level – e.g. national forest inventory systems (at least those that are reliable and systematic over time).
The direct involvement of countries is seen as critical to success. Countries will be sceptical about the utility/accuracy/relevance of a world forest survey at the national level if it operates independently of the country’s own inventory program.
A case needs to be made as to the cost to benefit ratio of a world forest survey. It has been seen as a supplement to FRA, but in the future may evolve to simply become the future FRA. It could also be viewed as simply an elaboration of the remote sensing component of FRA. A compilation of country inventory data will still be required.
It is very relevant for places like Africa where there is a lack of consistent data for many countries in the region. Africa might be a good place to initiate a pilot project.
The world forest survey concept is very consistent with an increasing set of global accords (e.g. Kyoto, Montreal Criteria and Indicators, Helsinki accord, etc.) regarding global forest sustainability. It could provide a framework for addressing the various criteria in a consistent and meaningful fashion.
The system may require more than two levels (TM and ground). It may need a level or two in between (such as a higher resolution imagery sample) to increase efficiency. It is important to decide if the objective is a tree (cover) monitoring program or a forest (land use) monitoring program. A high proportion of trees may exist outside the forest. The expert team decided to continue as an ongoing peer review/advisory team for the FRA process, including possible evolution to a world forest survey.
