Previous Page Table of Contents Next Page


Chapter 9. Wood supply

ABSTRACT

Studies were made on the extent of forests theoretically accessible for industrial wood supply and on actual harvesting and wood removals by country. The study on accessibility was carried out using global maps and geographic information system (GIS) technology. A spatial model was developed to estimate forest area within varying distances from major transportation infrastructure, excluding protected areas and forest areas considered to be above an economically exploitable altitude. Adjustments were made for closed and open/fragmented forest classifications, differences among geographic regions and ecological zones, and distortions due to map projection. It was estimated that 51 percent of the world's forests lie within 10 km, and 75 percent within 40 km, of major transportation infrastructure. Results are presented by region and ecological domain. Boreal and tropical forests are more remote than other forests. In some regions, notably North America, protected areas represent a significant accessibility limitation. Protected areas have, however, a minor impact on accessibility at the global level. For the study on extent and intensity of forest harvesting, information was analysed for 43 tropical countries, representing approximately 90 percent of the world's tropical forests. About 11 million hectares of tropical forests were harvested annually, with intensity ranging from 1 to 34 m3 per hectare. Detailed information on removals and harvesting was assembled for industrialized countries. Excluding the Russian Federation, over 70 percent of the increment was harvested in these countries. Data are reported by country in Appendix 3.

INTRODUCTION

The world's forests present a large potential for industrial exploitation. Information on forest areas accessible for wood supply is important for land use planning, for development of forest industries and from a policy perspective. At the same time, accessibility can be seen as a potential threat to degradation of forest ecosystems. Two studies on wood supply were done in FRA 2000 - one to estimate the theoretical accessibility of forests for industrial wood supply, and one to collect information on areas and volumes actually harvested.

The study on accessibility was based on the FRA 2000 global maps on forest cover, protected areas and ecological zones. Because it was based on coarse-resolution data, the study is not useful at the local level; the study was intended as a reference for policy-makers at the national and international levels.

The concepts used in the study were developed in previous studies, including the Global Fibre Supply Model (GFSM) (FAO 1998) and the Temperate and Boreal Forest Resources Assessment (TBFRA) (UNECE/FAO 2000). Both earlier studies included a compilation of recent forest inventory statistics. GFSM focused on access to global industrial fibre sources including non-wood fibre. Projections were made from inventory data to estimate the current situation as well as to forecast accessibility of raw material. The UNECE/FAO study was based on the results of a comprehensive survey of industrialized countries with temperate and boreal forests. Both studies reached similar conclusions to FRA 2000, with some variations for some geographic regions and countries.

The study on actual harvesting and removals was mainly based on information obtained from national reports. Most previous assessments of forest harvesting and removals have been restricted to case studies of individual operations, providing a mosaic of information on several topics under specific conditions. For FRA 2000 a more comprehensive overview was requested of the extent of forest harvesting schemes and intensity of harvesting by country.

METHODS

Accessibility

The study was based on an analysis of global thematic maps of forest cover, protected areas and global ecological zones produced for FRA 2000. The mapping processes are described in detail in Chapter 47. The analysis was carried out with ArcView 3.2 software using grid themes with cell size of 2 × 2 km2 in the Robinson world map projection with central meridian 0. The resulting figures were post-processed to correct area distortions due to the map projection.

Table 9-1. Correction factors for closed and open/fragmented forest by geographic region and ecological zone

Geographic region/ecological zone

Closed forests

Open/fragmented forests

World except North and Central America



Dry tropical and subtropical zones

3.0

2.0

All other ecological zones

1.0

0.5

North and Central America



All ecological zones except temperate and boreal United States and Canada

1.0

0.5

Temperate and boreal United States and Canada

0.7

0.3


The global ecological zone map (FAO 2000a) was developed by several institutions, including the United States Geological Survey (USGS) EROS Data Center (EDC); the Ecological Laboratory of Toulouse (LET), France; and the UNEP World Conservation Monitoring Centre (UNEP-WCMC). The base for the classification of the ecological zones was the Köppen-Trewartha system and the data set that describes five domains based on temperature: tropical, subtropical, temperate, boreal and polar, with a further division into a second level of 20 global ecological zones, as recognized by FRA 2000. Data are presented at 1 km ground resolution.

The forest cover map (FAO 2000b) was produced by FAO and EDC and comprises woody vegetation cover divided into three classes: closed forest, open/fragmented forest and other wooded lands. The data set was based mainly on remote sensing material and presented at a 1 km ground resolution.

The protected areas map (UNEP-WCMC 2000) was produced for FAO by UNEP-WCMC and consists of both international and national protected areas at 1 km ground resolution.

A digital elevation model was derived from two topographic data sets developed by EDC, HYDRO1k (EDC 1996b) and GTOPO30 (EDC 1996a), both with the resolution of 30 arc-second corresponding to approximately 1 km ground resolution.

As standard map data several themes were employed from the ESRI Digital chart of the world (ESRI 1995). The political theme was employed for country boundaries, the road and railroad themes were used to describe land transport infrastructure, and the inland water drainage theme was employed to describe water transport infrastructure.

The data sets were geometrically adjusted and all grid themes were resampled to the cell size of 2 × 2 km2 at the extension of the world in Robinson map projection to achieve a usable overlay of all themes and a feasible level of detail to perform the map analysis.

The following assumptions were made regarding the availability of forest for wood supply.

In tropical South America major rivers below 400 m altitude were considered as additional transportation routes. Rivers are used for wood transportation also in other parts of the world, but the road and railroad infrastructure already covers major parts of these regions. From the major transportation routes it is understood that forests are accessible through minor forest roads and streams.

Two major adjustments were applied to the area figures from the map analysis. First, adjustments were made for closed forest and open/fragmented forests in each geographic region and ecological zone. The correction factors were determined manually (Table 9-1), based on comparisons between the forest areas by region in the global map and the forest areas given in country reports. Dry forests tended to be underrepresented in the remote sensing results used to produce the forest cover map. Tropical and subtropical dry forests were therefore given a higher correction factor. The class open/fragmented forest in the forest cover map partly represents mosaics of forests, and the area figures for this class must therefore be adjusted downwards. In North and Central America the forest areas were over represented, especially in the United States and Canada, and therefore lower correction factors were used.

Second, adjustments were made for area distortions caused by the map projection. For practical reasons, all maps were processed in the Robinson map projection (Robinson 1963), which is one of the most frequently used projections for global maps and presents a "true" shape of the world. The projection presents a uniform scale only within latitudes ±38° and the scale error becomes greater as latitude increases. The areas derived from the Robinson map were calibrated for each country by calculating the relation between the areas in Robinson map projection and the areas in Lambert Azimuthal Equal Area map projection (a map projection that displays geographical data by continent at a fairly true scale). The calibrated statistics contained corrected proportions of areas.

Harvesting and removals

Tropical countries. In-depth analysis on removals and harvesting was carried out for 43 selected tropical countries where detailed country references were available. The countries included in the study represented the forested tropical countries of Africa (19 countries), Asia including Oceania (10 countries) and America (14 countries) (FAO 2000c). The selected countries have in common that their forest cover exceeds 5 percent of the total land area or exceeds 1 million hectares. Collectively, the 43 countries accounted for approximately 90 percent of the world's tropical forests.

Information was gathered both through an extensive literature review and through statistical modelling, supplemented by country visits to Gabon, Suriname and Papua New Guinea. The literature research focused on data published in government reports and on documentation where the original source could be determined. Analysed data included only commercial and legal utilization of the timber resource from natural broad-leaved forest in the tropics. Other forms of utilization of the resource (e.g. fuelwood, illegal or unrecorded harvesting) that might affect forest conditions in individual countries are mentioned in descriptive country profiles published in the FAO forestry country profiles (FAO 2001).

Industrialized countries. Industrialized countries generally have substantial information on harvesting which it was possible to collect for FRA 2000. Countries responded to detailed questionnaires covering several aspects of harvesting and removals, including breakdown by species and balance among increments, natural losses and fellings. A full account of the methods and results is found in UNECE/FAO (2000).

RESULTS

Accessibility

The results from the analysis by geographical region are presented in Table 9-2. These results are broken down also by ecological domain in Figure 9-1.

In Europe, access to wood supply in the vast areas of boreal forests is to a great extent limited by the absence of major transportation infrastructure. In contrast, in the European temperate and subtropical forests, the major limiting factor is protected area status.

Most of the forests in South America are found in the tropical region. Remoteness from roads and railroads in the Amazon basin implies the use of rivers as potential transportation routes. Water transportation improves the accessibility of wood supply areas by almost 50 percent (Figure 9-2), but accessibility still remains limited. The protection status of some forest areas is also a factor limiting access.

In North and Central America vast areas of boreal forests offer low accessibility for wood supply because of an undeveloped transportation infrastructure, while access to temperate and subtropical wood supply is relatively restricted by protected areas.

The tropical forests in Africa are generally closer to infrastructure than tropical forests in Asia and South America.

Accessibility to wood supply areas in Asia is mainly restricted by protected areas and altitude limits, and to a lesser extent by remoteness.

Tropical forest is the main forest type in Oceania, and its accessibility for wood supply is restricted by lack of transportation infrastructure. Protected areas are the major limiting factor for access to the region's subtropical and temperate forests.

Globally, 51 percent of the world's forests considered available for wood supply were found to be located within 10 km of major transportation infrastructure. Fourteen percent of the world's forests were considered unavailable for wood supply because they were located in protected areas or in areas above accessible altitude.

Table 9-2. Forest area at varying distances from major transportation infrastructure, and accessibility for wood supply, by region

Forest accessibility, by geographical region

Total forest area million ha

Cumulative forest areas at varying distances from major transportation infrastructure million ha

10 km

20 km

30 km

40 km

50 km

Unlimiteda

Africa

650







Inaccessible forest


n.a.

n.a.

n.a.

n.a.

n.a.

n.a.

Forest in protected areas


33

53

62

65

67

69

Forest available for wood supply


422

533

562

572

576

581

Proportion of forests available for wood supplyb


65%

82%

86%

88%

89%

89%

Asia

548







Inaccessible forest


18

24

25

26

26

26

Forest in protected areas


34

46

51

53

54

59

Forest available for wood supply


344

412

430

439

444

462

Proportion of forest available for wood supplyb


63%

75%

79%

80%

81%

84%

Oceania

198







Inaccessible forest


n.a.

n.a.

n.a.

n.a.

n.a.

n.a.

Forests in protected areas


10

15

17

18

18

21

Forest available for wood supply


110

141

153

159

164

177

Proportion of forest available for wood supplyb


56%

71%

77%

81%

83%

90%

Europe

1 039







Inaccessible forest


12

22

28

34

39

56

Forest in protected areas


18

23

26

29

31

37

Forest available for wood supply


518

657

727

776

813

946

Proportion of forest available for wood supplyb


50%

63%

70%

75%

78%

91%

North and Central America

549







Inaccessible forest


11

15

17

19

20

24

Forest in protected areas


61

82

88

92

94

101

Forest available for wood supply


248

309

335

351

363

424

Proportion of forest available for wood supplyb


45%

56%

61%

64%

66%

77%

South America

886







Inaccessible forest


1

2

2

2

2

2

Forests in protected areas


24

41

53

62

68

141

Forest available for wood supply


333

475

554

608

644

742

Proportion of forest available for wood supplyb


38%

54%

63%

69%

73%

84%

World

3 869







Inaccessible forest


41

62

73

81

87

109

Forests in protected areas


180

261

297

318

333

428

Forest available for wood supply


1 976

2 527

2 761

2 906

3 004

3 332

Proportion of forest available for wood supplyb


51%

65%

71%

75%

78%

86%

a Unlimited distance implies that all forests are within economic reach.
b Proportion of total forest in the region that is accessible for wood supply and within reach of transportation infrastructure.
The results from the analysis by ecological zones (Figure 9-1) indicate that the main limit to accessibility for wood supply in tropical forests is remoteness from transportation infrastructure, especially in the Amazon region.

The subtropical forests are relatively accessible through transportation infrastructure. About 10 percent are not accessible because of protected area status, and about 6 percent of the forests lie above the altitude limits set for the study.

Most of the temperate forests are accessible by transportation infrastructure. About 15 percent are excluded because of protected status.

The world's boreal forests are to a great extent limited for wood supply because of remoteness and to some extent also because of altitude and protected status.

Figure 9-1. Proportion of forests within reach from major transportation infrastructure and its accessibility for wood supply

Notes: Boreal forest also includes forest in the polar domain. Major rivers were only considered as transportation infrastructure in tropical South America (see assumptions). The area of forest by ecological zone accounts for all forests, in contrast with the analysis in Chapter 2 where countries as a whole were considered tropical.
Harvesting and removals

Tropical countries. Country information on harvested area, harvesting intensity and volume harvested for the 43 studied tropical countries is presented in Appendix 3, Table 10.

In Africa, of the 5.9 million hectares under timber harvesting schemes, 3.3 million hectares were harvested annually. The harvesting intensity was highly variable in the countries, ranging from 1 m3 per hectare in Zambia to 13 m3 per hectare in Gabon. In Asia and Oceania, the total forest area under a harvesting scheme in 2000 amounted to 27.3 million hectares, of which 6.2 million hectares were actually harvested each year. The harvesting intensity in most countries was higher than in Africa and ranged from 5 m3 per hectare in Myanmar to 23 m3 per hectare in Viet Nam. In tropical America, a total of 16.7 million hectares were under a timber harvesting scheme, while some 1.9 million ha were harvested annually. The harvesting intensity ranged from 1 m3 per hectare in Bolivia to 34 m3 per hectare in Brazil.

Figure 9-2. Proportion of forest areas available for wood supply within reach of land and complementary water transportation infrastructure in tropical South America

Note: For the other countries in tropical South America the water infrastructure option did not make any significant additional forest areas available for wood supply.
Industrialized countries. As reported in TBFRA (UNECE/FAO 2000), annual fellings in the temperate and boreal domains were 1 632 million cubic metres in the mid-1990s. Of this, over half, 922 million cubic metres, was in the two countries of North America, and another 28 percent in Europe (465 million cubic metres). The Russian Federation, which accounts for 30 percent of the region's net annual increment, accounted for only 9 percent of its fellings. The contrast between growth and harvest in the Russian Federation was linked to the country's economic, social and infrastructure problems related to the process of economic transition. In the mid-1980s, Russian fellings and removals were at least three times higher than at the end of the 1990s. This dramatic decline in fellings of the world's largest forest resource has significant consequences for global wood supply and the global carbon balance.

Removals from temperate and boreal forests were 1 260 million cubic metres, implying considerable harvest losses for the region as a whole. Taking into account only forest land, removals were 88 percent of fellings in Europe, 74 percent in the Commonwealth of Independent States (CIS) and 90 percent in North America. Much of this variation may be explained by differences in conditions and practices among regions.

For the temperate and boreal domains as a whole, fellings of growing stock accounted for 53 percent of net annual increment. There are, of course, rather wide differences between regions. In general, a larger portion of the increment is harvested in those regions with powerful forest industries. Thus the proportion is 79 percent in North America, 72 percent in the Nordic countries and 63 percent in central-western and northwestern Europe. In the CIS it is only 17 percent, while in other countries of the temperate and boreal domain it is 52 percent.

If the data are broken down by species, it is apparent that coniferous stands are used much more intensively than broadleaves. The felling/increment ratio was 62 percent for conifers and 42 percent for broadleaves. For Europe, the ratio was 68 percent for conifers and 56 percent for broadleaves. In North America, the ratio was 98 percent for conifers and 54 percent for broadleaves.

CONCLUSIONS

Accessibility

The study on accessibility of forests for wood supply showed that 51 percent of the world's forests were both available for wood supply and located within a distance of 10 km from major transportation infrastructure. Fourteen percent of total world forests were considered unavailable owing to their location either in protected areas or above accessible altitude.

Boreal and tropical forests are in general more remote than the forests in other regions. This suggests that additional areas for wood supply will depend on the development of roads and railroads in such remote areas.

Regionally, protected areas represent a significant limitation for access to wood supply areas, for example in the subtropical and temperate forests in the United States and Western Europe. Note that some protected areas were not included in this study because of missing information on spatial extension (particularly for some European countries), in which case the forest area accessible for wood supply may have been overestimated.

It should be noted that there are some differences in the findings of this study when compared with the UNECE/FAO (2000) study of industrialized countries, based on differences in assumptions and in methodology.

A major strength of this study is that the same model was used for the entire world, enabling global comparisons. The model used is flexible for changes in input data, which makes it possible to update the results as new data become available. Map analyses were made at ground resolution of 2 × 2 km2, which resulted in high-quality output. The original data held a resolution of 1 × 1 km2 but it was not feasible to maintain that level in the digital map material because of higher processing requirements. All map sources were based on recent detailed remote sensing techniques and inventories and validated by various institutions and international experts.

Some protected areas in the UNEP-WCMC database are only recorded as point data without further spatial specification; these were not included in the analysis, since geographic representation is essential for the GIS analysis. If point data had been included in the analysis as circle areas (as was done in the protected area statistics elsewhere in FRA 2000), the total forest area under protection would have been about 20 percent higher. On the other hand, some forests in protected areas may be available for wood supply under certain conditions that comply with the management practices allowed in the protected area. These potentially offsetting factors are not further elaborated in this global study.

Harvesting and removals

Incomplete data on timber removals and harvesting in tropical countries made it difficult to reach global or regional conclusions. Studies carried out under FRA 2000 in tropical countries found a very wide range of harvesting practices and intensities. The information assembled for FRA 2000, however, constitutes a first attempt to establish a baseline by country.

For industrialized countries, more data were available. However, some inherent quality problems were almost impossible to resolve, and must be kept in mind when the data are used. Data were mostly supplied by national forest inventories and based on measurements in the forest, taken over the life cycle of the forest inventory. They are therefore not necessarily comparable with the annual data on removals published at the national level and in the Timber Bulletin and the FAO Yearbook of Forest Products. The latter are often estimates based on parameters that are more easily measurable on an annual basis, such as inputs of raw material to the forest industries. From time to time these estimates for annual felling are calibrated against the forest inventory data. For this reason, TBFRA data should not be directly compared to annual removals data published elsewhere, including other data from UNECE/FAO.

If the 10 km distance to infrastructure is taken as a reference, about half or just under 2 billion hectares of the world's forests are accessible for wood supply, distributed fairly evenly between tropical/subtropical and temperate/boreal domains. In the tropical domain, about 11 million hectares were harvested annually, which represents about 1 percent of the accessible area. Area data for removals were not reported for the temperate and boreal domain, but (excluding the Russian Federation) the removals are above 70 percent of the increment, indicating a higher intensity of wood extraction for industrial purposes than in the tropical domain.

BIBLIOGRAPHY

EROS Data Center (EDC). 1996a. GTOPO30. United States Geological Survey.
http://edcdaac.usgs.gov/gtopo30/gtopo30.html

EDC. 1996b. HYDRO1k. United States Geological Survey.
http://edcdaac.usgs.gov/gtopo30/hydro/index.html

ESRI. 1995. Digital chart of the world.
http://gisstore.esri.com/acb/showdetl.cfm?&Product_ID=313

FAO. 1998. Global fibre supply model. Rome.

FAO. 2000a. Global ecological zone map. Rome.
www.fao.org/forestry/fo/fra/index.jsp

FAO. 2000b. Forest cover map. Rome.
www.fao.org/forestry/fo/fra/index.jsp

FAO. 2000c. Environmental impact assessment related to forest utilization. FAO/ Government Cooperative Programme/Austria. Project Report. (unpublished)

FAO. 2001. Reduced impact logging in tropical forests: literature synthesis, analysis and prototype statistical framework. Working Paper FOP/08. Rome.

Robinson, A., et al. 1978. Elements of cartography. New York, John Wiley. 4th edition.

UNECE/FAO. 2000. Forest resources of Europe, CIS, North America, Australia, Japan and New Zealand: contribution to the global Forest Resources Assessment 2000. Geneva Timber and Forest Study Papers No. 17. New York and Geneva, UN.
www.unece.org/trade/timber/fra/pdf/contents.htm

UNEP-WCMC. 2000. WCMC protected areas map.
www.wcmc.org.uk/protected_areas/data


Previous Page Top of Page Next Page