Previous PageTable Of ContentsNext Page


Gaps in national-level information on forests and trees in developing countries

M. Saket

Mohamed Saket is a
Forestry Officer in the
FAO Forestry Department,
Rome.

Almost two-thirds of the developing countries were unable to provide satisfactorily detailed information to the Global Forest Resources Assessment 2000 (FRA 2000).

The Global Forest Resources Assess- ment 2000 (FRA 2000) was in- formative not only regarding the state of the world's forests, but also regarding the state of forest resources information. An analysis of the results indicates many areas where the country information is insufficient relative to the needs at the national level (where users include government institutions and civil society) and at the regional and international levels (where users include international agencies, multilateral development banks, donor agencies and non-governmental organizations with forest-related activities). This article reports on an evaluation of the information that developing countries provided to FRA 2000, revealing some important gaps.

1 - Percentage of all developing countries using various methods of data collection

2 - Percentage of forest area per data collection method, all developing countries

EVALUATION METHODS

The state of information on forest and tree resources in the developing countries was evaluated by reviewing the methodologies used by the countries to collect the information provided to FRA 2000 (Table 1; Figures 1 and 2).

Country-wide field sampling provides the most detailed and reliable information on a wide range of resource attributes. Countries that rely on mapping alone tend to have only general information on forest area, and no national-level information on volume and biomass in forests, in trees outside forests or on other wooded land. This information is then often extrapolated from partial surveys or obtained through expert guesses.

TABLE 1. Methods of collecting data on forest resources in developing countries                            

Method of data collection

Africa

 Asia

Latin America and the Caribbean

Oceania

Total

No. of countries

%

% of forest area

No. of  countries

%

% of forest area

No. of countries

%

% of forest area

No. of countries

%

 % of forest area

 No. of countries

 %

% of

       forest area

Detailed mapping and country-wide field sampling

 6

11

22

0

0

0

0

0

0

0

0

0

 6

4

7

Country-wide field sampling

13

23

17

1

3

 n.s.

0

0

0

0

0

0

14

9

5

Detailed mapping

7

13

 37

10

27

30

20

 44

96

5

28

94

42

27

62

General mapping

 6

11

 5

6

16

26

3

 7

 n.s.

0

0

0

15

10

7

Expert estimate

24

43

20

20

54

45

22

49

4

13

72

 6

79

51

18

Source: FAO, 2001.
n.s. = not significant.

 

Country-wide field sampling provides the most detailed and reliable information on a wide range of resource attributes; shown here, an inventory group in Costa Rica discussing technical details of field methodology

- M. SAKET

EVALUATION RESULTS

The general tendency in developing countries is to obtain forest information through the use of remote sensing for detailed mapping. Almost one-third of the countries (31 percent), which account for 70 percent of the total forest area in developing countries, made detailed forest cover maps (Figures 1 and 2). One-quarter of the developing countries (38 countries) - accounting for 64 percent of the developing countries' forest area and 34 percent of the world's forest area - updated their national remote-sensing mapping between 1983 and 1999.

Country-wide field sampling is still scarce in developing countries. Only 13 percent of the countries, containing about 12 percent of the forest area, have relatively detailed data sets on their forest resources obtained through country-wide field sampling. The thematic breadth of the surveys is highly variable. Field surveys have often focused on timber volume and tree species composition and to a lesser extent on biomass and other forest and tree attributes.

In Africa, 43 percent of the countries, accounting for 20 percent of the continent's forest area, supplied information to FRA 2000 based on expert estimates (Table 1). Country-wide field sampling together with detailed mapping covered 22 percent of the forest area, and country-wide field sampling alone covered 17 percent of the forests; thus detailed information on forest resources was reported for about 39 percent of the region's total forest. For 61 percent of the forest in the region the information was generally limited to forest area estimates obtained through general mapping and/or expert estimates.

Some 54 percent of the countries in Asia, accounting for 45 percent of the forest area in the region, have information based on expert estimates. Only one country reported that it carried out country-wide field sampling, and its area is too small in relation to the total forest in the region to be of significance. National information generated from country-wide field sampling together with detailed mapping was completely lacking. Thirty percent of the region's forest cover was mapped at the national level with sufficient detail for forestry purposes between 1990 and 1999. For nearly 100 percent of the forest in the region, information is generally limited to area estimates. Asian countries provided the poorest information on forest area, with 71 percent of the forest assessed through expert estimates or general mapping.

In Latin America, as in Asia, all countries obtain their forest information from remote sensing or expert estimates. Some 44 percent of countries in the region produced detailed forest maps, which covered nearly 96 percent of the forest area in the region. Latin America thus had the most reliable area information (followed by Oceania). Most of the remaining information was based on expert estimates. None of the countries has reported information based on national-level field sampling. Information from all countries is thus generally limited to area estimates.

In Oceania, 28 percent of the countries, which account for 94 percent of the forest cover, have information obtained from detailed mapping based on remote sensing work. Some 6 percent of the region's forest was assessed through expert estimates. Information for the whole region is thus generally limited to forest cover (or forest area) estimates.

Countries with less forest cover tend to invest less in forest information gathering based on field sampling or remote sensing. Baseline information for FRA 2000 showed that 94 countries out of 156 (61 percent) have no forest resources survey and use information produced through expert estimates or general mapping. These countries account for 47 percent of the total land area but only 25 percent of the forest area (Figure 2). National inventories based on field sampling were carried out in 20 countries (13 percent of the forest area); of these, only three of the countries (2 percent of the forest area) have repeated forest inventories. Meanwhile 48 countries (31 percent of the forest area) relied on detailed mapping to produce information of the extent of the forest cover.

CHANGE ESTIMATES

The number of reports on forest resources increased in the 1990s, and some countries notably improved their inventories by producing more reliable and up-to-date information on some parameters; 35 countries carried out new detailed mapping for area estimates, seven performed their first national inventories focusing on a larger range of parameters, and two updated their previous national inventories. Nevertheless, many countries still lack the basic data needed to assess the state and changes of their forest cover accurately.

A number of countries updated their forest cover estimates during the 1990s, often through mapping using remote sensing, but in many cases the methodology was not directly compatible with that of previous surveys, making change estimates difficult. Comparable multiple-date inventories are scarce, and there is a need to improve both the accuracy and depth of information provided in forest inventories. Almost one-third of the countries reported information for reference years of 1990 or before, heightening the risk of error in projecting estimates of forest cover to 2000 and estimating area changes for the period 1990-2000. Five countries did not report the reference year. The data reported in the last years of the 1990s were generally based on updating of historical information or on expert estimates.

Very few developing countries have developed information systems for their forests, and the difficulty of accessing information complicates the task of estimating forest cover change. In some countries information is still maintained in hard copies often difficult to locate and use.

OTHER PARAMETERS

In developing countries information on other forest and tree resource parameters besides area remains rare (Table 2). A total of 43 countries (28 percent) reported information on volume, biomass, species composition and protected areas. Systematic field inventories that measure volume, biomass, productivity of the forests and other parameters were carried out in many countries, but often for limited areas. As a result, national estimates for volume and biomass had to be extrapolated from local studies. Information on parameters related to biological diversity, forest management, protection status, forest fires, forest health, trees outside forests, total biomass, NWFPs, growth and removals was scarce and often of low reliability.

TABLE 2. Provision of forest information on various parametersa

Parameter

 Percentage
of developing countries that
provided information

Change in area

10

Growing stock

28

Biomass

28

Removals

26

Fires

4

Management

31b

Protection status

100c

Increment

Negligible

Ownership

Negligible

Regeneration

28

NWFPs

44

Species diversity

28

Forest health

Negligible

People living in or near forests

Negligible

Source: FAO, 2001.
a Information on other wooded land was lacking for most parameters.
b This figure includes six countries that provided partial information, with national figures not available.
c Developing countries were not asked to provide information on protected areas for FRA 2000. The figure of 100 percent reflects the coverage of the global map of protected areas provided to FRA 2000 by the United Nations Environment Programme's World Conservation Monitoring Centre (UNEP-WCMC), based on UNEP-WCMC's database of protected areas.

CONCLUSIONS AND RECOMMENDATIONS

In general, the lack of description of methods and techniques of data collection makes it difficult to assess the accuracy of national information properly. However, the FRA 2000 results showed that the availability of country information in the developing countries was not satisfactory for many subjects considered important for forest policy development. While a number of industrialized countries have reliable and continuously updated information on a large number of forest attributes, in the developing countries baseline information is still very fragmentary, particularly on other parameters besides forest area (e.g. volume, biomass, biodiversity and NWFPs), on other wooded land and on trees outside forests.

Reliability of country information is generally low. In responding to FRA 2000, 61 percent of the developing countries, accounting for 25 percent of the forest area in the developing regions, provided questionable information based on expert opinion or coarse and inaccurate mapping. From the entire set of parameters defined for global reporting to FRA 2000, countries produced reliable information only on forest area (for 69 percent of the forests). For 103 countries (66 percent of the developing countries' forest area), the information provided was of low to medium compatibility with previous country data (FAO, 2001).

Forest cover information for forests, other wooded land and other land (trees outside forests) was available for 69 percent of the forest area of developing countries, compared with 99 percent of industrialized countries' forest area (UNECE/FAO, 2000).

While many countries have invested in inventories of forest resources, little or nothing has been done to date to assess the tree resources on non-forest land. In some countries, particularly those with low forest cover, these resources are of great utility for their products and services. Governments generally recognize the multiple roles of trees outside forests and encourage their planting, but there is no sound statistical base to support policies and strategies for their conservation and development.

On the basis of the identified information gaps, the following recommendations can be made.

Mapping alone generally gives information on forest area only; information on trees outside the forest or other wooded land thus tends to be scarce and of low reliability

- M. SAKET

Bibliography


Previous PageTop Of PageNext Page