Thematic papers

Emerging trends in forests and forestry in the Asia-Pacific region and their implications on data collection and analysis of forest product statistics^[2]

Chris Brown^[3] and Chris Perley^[4]

Introduction

The Asia-Pacific region is the world’s most populous, housing more than 3.3 billion people - more than 55 percent of the global population. Its constituent countries are among the world’s most densely populated (for example Bangladesh, Singapore). Others such as Mongolia and Australia have the world’s lowest population densities. Countries like Japan and Brunei Darussalam are among the world’s richest nations, while Nepal and Cambodia are ranked with the poorer nations on the globe. The Cook Islands and the Solomon Islands have some of the highest proportionate forest cover in the world, while Pakistan and Tonga join other countries that have very sparse forest cover. China and Indonesia supply some of the world’s most extensive national forest areas, while Nauru and Tokelau are among the smallest countries, with the least forest area.

The forests of the Asia-Pacific make a vital contribution to the environment, societies and economies of the region. The region is strongly committed to the principles of SFM, although many countries are struggling to translate this commitment into meaningful action. Several countries in the region are leaders in implementing participatory forest management. Many countries have recognized that very high population pressures mean that effective SFM can only be implemented if communities and key stakeholders are strongly involved in forest management decision-making. At the same time, traditional cultural and bureaucratic structures are accountable for considerable inertia in the face of otherwise rapid change. Generally, institutions and groups that historically have held the rights to forest management have been loathe to surrender the economic benefits, power and prestige that accompany these responsibilities.

Consequently, the region is characterized by diversity - in cultures, in politics, in economies, in ecosystems and environments, and in the key national challenges to be faced. Dynamism and flux are functions of this diversity. Sources of diversity and change include the following:

• Different forest types, from moist tropical hardwood to boreal coniferous

• Continuing forest conversion and overexploitation

• Different forest uses/values, including:

  - industrial demand and community demand;
  - non-wood forest products (NWFPs); and
  - ecological services - biodiversity, natural hazard mitigation, etc.

• Diverse and dynamic social structures

• High population, with wide variation in density

• Diverse and dynamic political arenas - from stable to unstable (44 countries)

• Diverse and dynamic economic structures

• Emerging social and environmental trends:

  - international (e.g. Kyoto Protocol);
  - national (e.g. ITTO criteria and indicators for SFM); and
  - local initiatives (e.g. “ecosystem” management, public participation, devolution of responsibility).

Basic data on Asia-Pacific countries

Table 1 below provides data on land area, population, and economic indicators for each of the countries and subregions of the Asia-Pacific region.

Evaluation of forests in the Asia-Pacific region

The evaluation of both forests and forest product flows in the Asia-Pacific region is made more complex because of the dynamism and diversity of the region. For instance, projecting forest production from resource stocks and flows requires:

1. Net stocked areas, by owner and location.

2. Forest classification/crop-typing of forest quality.

3. Age classes, where appropriate to forest type.

4. Growth and yield data by product type and forest type.

5. Owners’ management assumptions (e.g. SFM).

6. Physical, ecological, social, political and economic constraints to production.

7. Where possible, reconciliation with demand.

Each aspect is subject to change, and requires continual assessment. Without knowledge of the detail, the evaluation and monitoring of policy options is constrained.

Forest cover change in the Asia-Pacific region

Table 2 provides details of the change in forest cover.

Table 2. Forest cover change in the Asia-Pacific Region, 1990-2000

In 2000, the total forest cover in the Asia-Pacific region was 698 million hectares. This represents 25 percent of the Asia-Pacific land cover, and 18 percent of the global forest area. Planted forest comprised 113 million hectares of the Asia-Pacific estate in 2000, an increase of 34 million hectares since 1990 (excluding Japan and Australasia). Most of this plantation increase occurred in India (increasing on average by 1.5 million hectares per annum) and China (1.2 million hectares per annum).

Natural forest constituted the remaining 585 million hectares in 2000, a reduction of 45 million hectares (-7 percent) from approximately 630 million hectares in 1990. This decline continues, with the current annual decrease estimated at 2.5 million hectares per annum. What is harder to assess is the qualitative decline, especially from an ecological point of view.

Why is there a decline in natural forest?

The decline in natural forest area is the result of three main drivers: forest fires, pests and diseases; land clearance for agriculture; and exploitative logging practices.

Forest fires generate either forest destruction (a quantitative decline), or degradation (a qualitative decline). Pests and diseases mainly affect forest quality. The yearly decrease from the result of forest fires is estimated at a very significant 10 million hectares. Some notable forest fires have captured worldwide attention, particularly the Australian bushfires of December 2001, the Indonesian fires of 2001 and the Mongolian fires of 2000.

Land clearance for agriculture is perhaps the most pervasive process that impacts on forest decline, resulting in a sometimes-permanent loss of area. The problem’s management and control is complicated by the influence of many economic and social drivers, especially high populations and poverty. The direct effects of logging are associated normally with forest decline. However, logging also provides access for those who want to convert land for agriculture.

Socio-political influences on future forest management

Policy developments

The development of national forest policies generally centres on the needs and requirements for the implementation of SFM. All Asia-Pacific Forestry Commission (APFC) countries recognize the need to advance towards the implementation of SFM, and the need to ensure that forests fully contribute to national development and benefit local communities. There is a high degree of commonality in the general thrusts of most national forest polices, although specific national situations dictate the overall policy emphasis in individual countries. Topics in forest policy development include deforestation and forest degradation, illegal logging, land-use planning, devolution and decentralization of forest management, and conservation.

Relevant policies, legislation and planning initiatives have included the following:

• The use of incentives to further SFM goals.

• A change of those policies that had increased forest loss; for example industrial and agricultural policies.

• Direct forest protection and logging bans.

Sustainable forest management

Sustainable forest management initiatives have focused on how SFM is to be achieved, particularly the structure of management decision-making processes. To this end, there has been a rise in nationally developed criteria and indicators (C&I) for SFM, as well as regional initiatives such as reduced impact logging (RIL). Internationally, the forest certification processes (such as the Forest Stewardship Council), use the same principles (or criteria) and indicators approach covering adherence to law; consideration and participation of local people; a minimum standard of sustainable yield with a recognition of the multiple values from forests (aside from timber production); and the detailed consideration of forest ecological values in the management of forests. The rise in public participation is also manifested in government policies and practised independent of C&I for SFM initiatives.

Forest institution restructurings

Rapid changes in forestry in the Asia-Pacific region are reflected in rapid changes in forestry institutional structures. A general shift in the region, towards more participatory philosophies, means that forestry departments in many countries have devolved significant forest management responsibilities. The challenge for these institutions is to reposition themselves to remain effective institutions, while ensuring staff acquire new skills to carry out significantly different types of work.

Governments throughout Asia and the Pacific are attempting to achieve SFM and biodiversity conservation by decentralizing authority and responsibility for resource management. Discernible shifts can be seen that move away from centralized forestry departments and towards the devolution of forest management responsibilities, to state or provincial agencies, to the private sector, and to community and NGO groups. The trend to decentralize is driven by a range of factors, including efforts to reduce central bureaucracies and cut budgets, some history of government forest management failures, growing commitment to more socially equitable forest management, and, in some cases, the international trend towards increased economic liberalization and market orientation.

Forest production in the Asia-Pacific region

Total roundwood production in the Asia-Pacific region in 2000^[5], amounted to 758 million m³, or 24.7 percent of global production. Production of wood for fuel in the region totalled 515 million m³, amounting to 34 percent of global fuelwood production. Roundwood production for industrial purposes amounted to 243 million m³, or 16 percent of world production. Coniferous industrial roundwood production in the Asia-Pacific region totalled 113 million m³, and non-coniferous production was 130 million m³. Industrial roundwood production from forest plantations in the region in 2000 was estimated to be 142 million m³, which suggests that current production from natural forests is somewhere in the vicinity of 100 million m³.

There remains considerable scope for improvement in forest-harvesting statistics. Some critics suggest actual harvesting in some countries may be twice as high as official figures due to reporting discrepancies, inefficient data collection, illegal logging and various other forms of corruption that result in distortion in the data.

Changes in industrial roundwood harvest

Figure 1 shows changes in industrial roundwood harvest levels in subregions during the past decade. Among the most evident features in the chart are the significant declines in harvest in South Asia and Insular Southeast Asia, largely as a result of declining harvests in India, Malaysia and Indonesia. Industrial roundwood harvest in these three countries is reported to have declined collectively by slightly more than 50 million m³ in 2001 compared with 1991 levels. Reduced harvesting in these countries is primarily a result of tighter regulations on forest harvesting, including a ban on harvesting in India’s natural forests. Changes in trade dynamics include import access in India, and shifts toward the production of higher value products. As noted above, at least part of the decline is a result of harvesting not being recorded in official statistics.

Figure 1. Changes in industrial roundwood harvest by Asia-Pacific subregion

The Advanced Industrialized Countries grouping has been separated to display Australasia (Australia and New Zealand) and Japan as separate components, to show the marked increase in Australasian harvests and a commensurate decline in Japan’s harvest.

The decline in Japanese roundwood production is primarily due to increasing costs rendering harvesting uneconomic in many areas. The Australasian industrial roundwood harvest has increased by 12 million m³ per annum during the past decade, and a similar magnitude of increase is expected during the next decade as new plantations mature. Similarly, the timber harvest in China increased modestly from 1991 to 2001, but this does not’ reflect a significant substitution of plantation grown timber for timber sourced from natural forests. During the next decade, plantation timber supplies in China are expected to increase dramatically.

Summary of trends

The Asia-Pacific region is characterized by the following trends:

• A decreasing quantity of natural forest

• Degradation in the quality of natural forest

• An increase in planted forest

• The emergence of environmental issues (“‘ecological sustainability”’), involving:

  - development and implementation of national C&I for SFM; and
  - implementation of policies that reduce unsustainable practices.

• The emergence of social equity issues (especially for local and indigenous people), involving:

  - a general devolution of responsibility, and decentralization;
  - a significant rise in public (and, more especially, local) participation; and
  - the emergence of “ecosystem” management.

• A dramatic change in industrial roundwood harvest, with the emerging significance of plantation production

Implications for forest statisticians

Obviously, forest statistics’ collection is affected by both the diversity and changes expressed above. The effects relate to:

• The quality and quantity of the forest estate, with quality concerning forest type, ownership, structural and compositional dimensions and presence of forest services.

• The existence and qualification of the emerging new forest values associated with the increasing participation of local people living adjacent to, or within, forests.

• Forest products, particularly the changing production and trade patterns of timber and emerging NWFPs.

Given these challenges of diversity and change, which are impacted by a wide range of drivers affecting forest production, the forest statisticians and policy analysts will require:

• Intensive, standardized and adaptable information collection and compilation systems.

• The capacity for both qualitative and quantitative analysis.

• The capacity to understand and continually monitor the underlying environmental, political, economic and social trends, the existence of which provides many of the base assumptions grounding data collection, data compilation, and data analysis in reality.

Improving forest product information flow in the context of globalization

Rohan Nelson^[6] and Peter Kanowski^[7]

Forest product information flow

Forest product information provides data that assist decision-making in:

• Management

• Investment

• Policy-making

There are various levels of information and decision-making, including:

• Global/international

• National/subnational

• Sectoral

• Individual entities

These decisions affect:

• Both the public and private sectors

Contentions relating to forest product information flow

• Issues at the international level mirror those nationally

• National situations characterized by:

  - fragmented and disjunctive efforts (disparate historical origins?);
  - insecure and diminished resourcing; and
  - information being data-driven rather than market-driven.

• All parties benefit from better information flows

Responses to these contentions

• Review real needs - ask users

• Focus on minimum core data sets

• Identify the most cost-efficient collection methods (i.e. recognize comparative advantage)

• Facilitate access to data by all parties

Challenging issues

• “Information is power” - with implications for confidentiality mechanisms

• “Time is money” - with implications for providing timely information

• “User pays” - which may be affordable for large corporations, but difficult for:

  - governments in a public good role; and
  - small enterprises.

• “Money talks” - affecting the relative importance of the forest sector

• “People matter” - on which data collection, analysis and dissemination capacity are dependent

Responses to these challenges

• Build more effective partnerships within and between governments, businesses, research institutions

• Build capacity development into operations

• Ensure there are realistic expectations

Roundwood removals in New Zealand: Where does all the wood go?

Judith Dennis^[8]

Introduction

This paper defines the forest products used to determine the roundwood statistics in New Zealand, discusses the methodology and identifies some of the limitations and potential errors of the data.

Definition

Roundwood is the term used to describe the volume of a length of a felled tree, with or without bark, and generally having a round cross-section, such as a log or post. It may include larger diameter branches, but in plantation forest management, as practised in New Zealand, the merchantable timber in these branches is minimal.

In New Zealand, we use “inside bark” (i.e. without bark), and exclude wood left behind in the forest (e.g. stumps, branches, other residues).

Why do we measure roundwood removals?

Roundwood removal estimates are an important measure of forestry activity in New Zealand. They are used by Statistics New Zealand as input to calculations of quarterly movements in Gross Domestic Product; the New Zealand Treasury for monitoring economic performance for the forestry and wood-processing sector; and by the Ministry of Agriculture and Forestry (MAF) as an input to macroeconomic forecasts.

Roundwood removal estimates also provide a verification of harvesting figures obtained from MAF’s National Exotic Forest Description (NEFD) survey, which is in effect an inventory of New Zealand’s planted production forests.

How do we measure roundwood removals?

Roundwood removals from New Zealand’s forests are an indirect estimate based on several data sources. These include MAF’s wood-processing surveys and export data produced by Statistics New Zealand.

MAF’s production surveys (conducted quarterly and annually) provide the volume of sawntimber, veneer, plywood, fibreboard, particleboard, pulp and paper produced in New Zealand. Calculating roundwood removals involves “working backwards” from a production volume to estimate how much wood has been harvested in order to produce this volume (that is from output to input). The annual surveys ask sawmillers and wood processors for data that will aid in these calculations. In addition, approximately a third of the logs produced from New Zealand’s forests are exported unprocessed, and there is a small component of wood chips produced directly from roundwood.

A log flow diagram showing results from the 2001 annual surveys is provided in Figure 1.

Figure 1. Log flow 2001 (values in million m³ per annum)

The components of the measurement of roundwood removals in 2001 were:

Sawlogs

Sawlogs are those logs that are fed into sawmills for sawntimber production and tend to be the better quality logs.

In 1976 the New Zealand Forest Service derived conversion factors for a variety of forestry products. The sawntimber conversion factors were:

• 2.0 for indigenous (now termed “natural”) species; and

• 2.2 for exotic (now termed “planted production”) species.

That is, 2.0 m³ of natural species of roundwood or 2.2 m³ of planted production species of roundwood are required for the production of 1.0 m³ of sawntimber.

However, since 1976 sawmill technologies have improved and the characteristics of the crop being harvested have changed. Therefore, 1976 conversion factors are no longer appropriate.

As part of the “Annual Survey of Sawmills and Chipmills”, mill owners are asked to estimate conversion factors for their mills. The conversion factor data have been collected annually since 1990.

In my 1996 study, individual mill data were analyzed graphically and with summary statistics by region and by mill size to assess normality and to identify outliers. It was decided that it would be appropriate to eliminate conversion factors greater than 0.8. (It was considered that mills supplying data of this order were producing flitches rather than sawntimber.) While there were problems with normality, the size of the sample - several hundred mills in each year - would minimize problems resulting from this lack of normality.

The results of the study are outlined in Table 1.

The above figures have 95 percent confidence with the following assumptions:

• One single sample conversion factor for each mill - a more precise estimate might be available if there was information on sampling variation within mills.

• The data is distributed normally - the boxplots produced suggest this is not always the case, although the sample size for each year overall is sufficiently large.

• The data is collected without error - this is the most problematic, but since the number of mills surveyed is so large, confidence is improved.

Overall, the data used has industry acceptance. The indigenous component needs more work, but indigenous sawntimber makes up less than 1 percent of the total sawntimber.

The data collection process has continued since 1996, and the data have been analysed each year to monitor changes. However, problems with the wording of the relevant question have been identified, and so this section has changed several times, which may affect the consistency of the resulting conversion factors. Because of these concerns, the factors were not revised each year. In 1997 the indigenous factor was revised slightly to 1.91, but subsequent changes to the survey for indigenous sawntimber have meant that less conversion factor data are available now for analysis. For planted production timber, the 2001 figure was revised downwards to 1.89. This downward revision runs against the trend previously identified, and needs further evaluation.

Export logs

These are unprocessed logs, exported directly. No conversion factors are applied, and the data come from the Statistics New Zealand’s (SNZ) INFOS database. The source of the data is New Zealand Customs. In the year ending March 2001, 32 percent of logs harvested from New Zealand forests were exported whole.

Pulp logs

Pulp logs tend to be small and younger logs that are chipped for producing pulp and, ultimately, newsprint and other paper products. They include a component of hardwood (usually eucalyptus). Again the log input is obtained from the pulp mills in the annual survey. Errors arise in that some mills provide tonnage of logs rather than cubic metres. It is also possible that the figure supplied may include all logs purchased by the mill in the reporting period, rather than logs processed. While a lack of consistency in responses from New Zealand’s seven pulp or paper mills can occur, good relationships with the major producers mean that problems in the data as supplied can be rectified. Clearly, we need to clarify the relevant question in the survey. It is also important to maintain relationships with producers.

Small logs

This category includes logs used in fibreboard and particleboard production, and also logs used for posts and poles. For fibreboard and particleboard, the log input is collected from the processors in the annual survey. These products also include a component of residues or chips, but this is not estimated, since it has been captured already, primarily from the sawlog estimates. The same error possibilities as identified for pulp logs apply here. In addition, we have no way of obtaining a precise estimate of posts and poles. Earlier work had concluded this figure to be stable at around 320 000 m³ annually. In the last year, however, this has been increased to 400 000 m³, in recognition of anecdotal evidence that farmers are upgrading their fencing after some good years.

Peeler logs

Peeler logs are used in the production of veneer, and then further processed into plywood or laminated veneer lumber (LVL). Only a small proportion of veneer produced in New Zealand’s 10 veneer mills is exported as veneer (approximately 6 percent in the year ending March 2001). Because there are only a few veneer plants, they can all be surveyed easily, and it is not necessary to calculate a conversion factor. The annual survey asks millers for their total log input, which is used for the roundwood removal estimates. Apart from the errors already discussed for pulp logs and small logs, this component of roundwood removals is considered reliable.

Export wood chips

As mentioned earlier, wood chips, as a by-product of sawmills, are used in the production of some panel products and in pulp production. However, some chipmills, and increasingly some sawmills, are chipping logs for direct export. The calculation of the whole log equivalent for this product is problematic. There appear to be huge inconsistencies in the units millers use to supply us with this data. It is often difficult to determine whether the data supplied were actually in green tonnes, oven-dry tonnes or bone dry units (BDU), although the question in the survey clearly asks for oven-dry tonnes. We also need to include an estimate of chips produced directly from roundwood for export, rather than as a by-product of the sawntimber process - this is to avoid double-counting of sawlogs. Our data seem to indicate this figure is about 38 to 39 percent. Because not all sawmills produce wood chips, either for export or further processing in pulp or panel mills, we do not have the “safety” of large numbers in our estimates. The method for estimating roundwood contributing to export chips involves the calculation of a conversion factor, incorporating an allowance to avoid double-counting, and applying this to export chip data from SNZ. However, this data seems subject to error. On occasion we have discovered, just from industry knowledge, that whole shiploads are omitted from specific ports’ data. There is also some anecdotal evidence that the information supplied is in tonnes rather than the stated BDU. Fortunately, this is a relatively small component of total roundwood (estimated at about 1.5 percent in the March 2001 year).

Figure 2 shows the contribution of the various log types to estimated roundwood removals.

Other uses of roundwood removals - roundwood equivalents

Earlier it was stated that information on log inputs supplied by mills is sufficient to provide data on roundwood inputs into panel products and pulp and paper products. However, analysts are interested in having a conversion factor for each of these products, particular those forestry analysts looking at added-value for forestry products.

This problem is not straightforward. The methodology for sawntimber is not usually applicable, in that most other forestry products include a residue factor from sawmills (wood chips, slabs, etc). The exception is veneer.

Figure 2. Estimated roundwood removals from forests in New Zealand

Veneer

The results of calculating a weighted conversion factor from the annual surveys for veneer mills since 1990 are provided in Table 2.

The improved conversion factor from 1993 onwards reflects the increased production and the known plant upgrades through the 1990s. The fluctuations from 1995 to 2000 could well be due to the measurement error referred to earlier. The largest producer reporting an unlikely change unduly affects the drop in the 2001 figure.

Plywood

Weighted conversion factors have been calculated since 1990. The methodology is similar to that used for veneer, with those mills producing plywood or LVL being asked for the amount of their own veneer used in further production. (MAF’s forestry statistics have always included LVL in plywood data, for reasons of confidentiality.) The weighted conversion factors are provided in Table 3.

These figures are more variable than those for veneer, although the general improvement from the mid-1990s is visible. Again, there is a marked divergence for the 2001 data. The survey questions for this information in the panel products’ questionnaires have not changed markedly over time, and it is difficult to understand the discrepancies both between mills each year and across mills over time.

It is important to understand that discussing conversion factors for plywood is not sensible as a component of log-flow usage, since the logs have already been “counted” as peeler logs in the production of veneer.

Other forestry products

Fibreboard (mostly medium density fibreboard - MDF - but some hardboard and softboard): A wood-based composite material that uses wood fibres, rather than particles or veneers to produce board or sheet products. Logs are either chipped at the fibreboard plant, or chipped material is purchased from chipmills or as residue from sawmills. (The “double-counting” problem can arise here in relation to residues from sawmills.) Historically, information collected in the fibreboard production survey has not produced good quality information for calculating conversion factors.

Particle board: Any board that is made from wood particles, as opposed to wood fibres or sheets of veneer and includes particle board and oriented strand boards (OSB). These products are primarily used in the construction market (e.g. wall panelling and flooring). Particle board was developed to make use of wood-processing waste, such as planer shavings and sawmill slabwood, offcuts and sawdust. Consequently, it is not possible to calculate a conversion factor directly.

Pulp: Logs are either chipped at the pulp mill or chips are purchased from other plants. The direct calculation of a conversion factor is not straightforward, particularly as the amount of wood required is different for chemical pulp and mechanical pulp.

Since paper is made directly from pulp, any attempt to convert log input to paper production is a double-counting. However, it is still interesting to estimate the amount of wood used in paper production.

MAF’s Statistical Releases on Exports of forestry products include an attempt to take into account the use of residues in calculating a roundwood equivalent figure to measure the total wood exported annually. However, this should be regarded as an estimate at best.

Validating harvesting data

One of the purposes of calculating roundwood removals identified earlier was to validate harvesting data from the NEFD survey. Comparisons with the NEFD are shown in Table 4.

Further work

While the estimated roundwood removals calculated from the forest products are close to the NEFD survey harvest volumes there is potential to improve the roundwood estimate through the following work:

• Clearer definitions and “tightening” of the questionnaire

• Maintaining and potentially improving the quality of the export data through liaison with Statistics New Zealand

• Providing more precise regional estimates

• The development of a more consistent methodology