Part II: National forest monitoring systems – purposes, options and status

This text is reproduced with the kind permission of CAB International. Original reference:

Holmgren, P. & Marklund, L-G. 2007. National Forest Monitoring Systems - purposes, options and status. In: Forestry & Climate Change edited by P.H. Freer-Smith, M.S.J. Broadmeadow & J.M. Lynch, pp 163-173. CAB International. ISBN 139781845932947.

1 Introduction

Forests and forestry are subjects to a variety of political processes and high-level decision making that affect literally everyone’s habitat and/or livelihood. But do we actually know what we need to know for dealing soundly with forests and forestry? Are there monitoring processes in place to ensure that sufficient knowledge about forests and forestry is generated to reduce uncertainties and support wise decisions?

Over the past decades, at least ten legal international instruments were established that address specific aspects of forest resources, their management and uses (Ruis 2001). Efforts to agree on an overall forest convention, on the other hand, failed at the United Nation’s Conference on Environment and Development (UNCED) in 1992. Although alternative explanations of this failure exist (Davenport 2005), it is commonly acknowledged that the main argument against a forest convention was – and remains – the protection of national sovereignty. Many countries wanted to retain governance over national forest resources and not surrender decision-making to a binding international agreement.

The UNCED negotiations thereby reinforced the national-level nature of forest policy and legislation. Over the following fourteen years, the international dialogue has not closed in on any international and legally binding arrangement on forests (Persson 2005). A starting point for this paper is, consequently, that forest policy and legislation are inherently national-level processes, and further that these processes depend on some form of national forest monitoring systems that meet defined requirements for the necessary decision-making.

At the same time, the numerous forest-related international agreements that have been successfully established require parties (countries) to verify and substantiate their compliance through specified reporting arrangements (CPF 2006). One key agreement relating to forests and forestry is the United Nations Framework Convention on Climate Change (UNFCCC) where a current debate concerns potential avoidance of carbon emissions from deforestation and from forest degradation. While interpretation and application of these concepts are to be negotiated, it is clear that national forest monitoring systems must be in place to follow and document the extent to which parties have followed the agreed intentions.

The question examined in this paper is whether current national forest monitoring systems, or national forest inventories, respond to requirements from national policy decisions as well as international reporting requirements. National forest monitoring systems are defined for this paper as processes that support strategic decision-making by:

• systematically and repeatedly measuring and observing forest resources, their management, uses and users;

• periodically delivering valid, representative and relevant information on status and trends for the country as a whole.

For the purpose of this paper, monitoring of operational forest management, including of legal compliance, early warning systems for example forest fires, of value-adding processing beyond the forest gate and of forest products are not included.

The paper determines the purposes of national forest monitoring. It evaluates technical options and approaches for monitoring systems against these aims. Finally, the situation of national forest monitoring is reviewed with special reference to the current discussion on avoiding carbon emissions from deforestation and from forest degradation within the UNFCCC.

2 Purposes of national forest monitoring

The purposes of national forest monitoring can be defined by objectives as expressed by relevant policy processes, under the assumption that such objectives also, implicitly, express a demand for systematic and quality-controlled information. Examining forest policy documents, including national communications to the United Nations Forum on Forests (CPF 2006), national forest programme updates (FAO 2006a), regional reviews (e.g. FAO 1998), and global overviews (e.g. FAO 2005), makes clear that forest policies are concerned with socio-cultural, economic as well as environmental dimensions (Table 1). Further, it is often emphasized that the forest sector interacts with several other sectors, e.g. agriculture, energy, tourism and transport.

Table 1. Issues frequently mentioned in recent expressions of forest policy objectives

Dimension	Issues
Socio-cultural	Rural livelihoods, Indigenous people’s rights, Rights of access, Tenure and land ownership
Economic	Poverty, Food security, Wood productivity and supply, Valuation of forest products and services, Equity, Trade, Energy
Environmental	Biological diversity, Soil and water protection, Climate change, Desertification, Air pollution, Invasive species, Wildfire, Pests

Sources: CPF (2006), FAO (1998, 2005, 2006a)

The broad set of forest policy objectives at national level resonates well with statements in the international dialogue on forests. Sustainable forest management, an umbrella concept for forestry ambitions, has been defined by seven themes (FAO 2006b) which also contribute to overall sustainable development aspirations. The framework of sustainable forest management has been used to define information requirements for the Global Forest Resources Assessment process, in which countries report based on their national information sources (FAO 2006c).

Therefore, national forest monitoring systems need to be designed to deliver cost-effective and quality-controlled information across the issues listed in Table 1. These systems must include a wide range of variables addressing biophysical as well as socio-cultural and economic issues. In response to these requirements, the scope of the national forest inventory concept has recently been expanded from traditional measurements of trees and other biological features, to include also interviews with local forest managers and stakeholders. (Kleinn et al. 2005, FAO 2006d).

Other emerging foci of forest monitoring relates to climate change. On one hand the health of forests under potentially changing climate need to be monitor for informed decisions and guidelines on adaptation of forest management practises. On the other, and as stated above, the mitigation of climate change through forest management by storing carbon in the forest ecosystem may become an economic and financial tool for forestry. Monitoring of carbon storage in the forest is closely related to variables covered by conventional national forest inventories, such as growing stock, growth and yield and forest area.

3 Options for national forest monitoring

This overview of technical options for national forest monitoring covers (a) methods for data collection, and (b) approaches to the national monitoring task.

Data collection methods

The wide range of social, economic and environmental issues to be addressed, imply a similarly wide variety of variables to be observed. Cost-effective methods for data collection have been a major focus in forest monitoring research (e.g. Ranneby et al. 1987, Gillis et al. 2005, Kleinn et al. 2005). Balancing requirements on accuracy and precision for the monitored variables with the cost of obtaining data from the field poses a classical problem of forest inventory. Cost may be prohibitive for systematic observations of some variables, e.g. soil carbon content. For others, key shortcuts include (a) applying statistical sampling – out of a wide variety of existing methods, (b) using subjective observations rather than more costly measurements, and (c) remote sensing to reduce the need for field work. Further, some variables are not directly observable as they reflect human perceptions and values, so data have to be collected through interviews with local stakeholders. Table 2 summarizes basic methods of data collection that are used in national forest monitoring systems. It is well established that a combination of these methods is required to monitor the range of identified forest and forestry issues at the national level.

Table 2. Basic data collection methods for national forest monitoring systems

Source: Builds on Holmgren & Thuresson (1998), Kleinn (2006) and Andersson (2006).

Approaches

National forest monitoring systems, as defined above, should systematically and repeatedly measure and observe forest resources, their management, uses and users. However, many countries have not established such systems, or have systems that only partially can be characterized as a national forest monitoring system (FAO 2006c).

The design and implementation of national monitoring of forests can be seen as an investment in information that pays off through increased future benefits to society accruing from forest resources. However, these increased benefits, and thereby the return of the investment, are often not well known and can not be easily generalized. The monitoring approach and ambition chosen by a country will depend on many internal factors that are not analysed here.

Instead, a generic range of available approaches, currently in use, has been developed for this paper (Table 3), based on country reports to the Global Forest Resources Assessment 2005. While no comprehensive evaluations of these approaches are made here, Table 3 indicates a measure of quality and reliability in the derived information, with the highest quality and reliability at the top and the lowest at the bottom.

Table 3. Optional approaches to generate national forest information, with examples from sources for the Global Forest Resources Assessment 2005

Approach	Typical properties	Pros	Cons	Examples (FAO 2006c, 2006d)
1. National forest inventory
Traditional national forest inventory	Many (>10000) systematically sampled field plots; Focus on biophysical variables	High precision and accuracy	High cost; Long implementation; Normally limited to biophysical variables	Sweden Finland
FAO-supported national forest assessment	Few (150-500) systematically sampled tracts; Measurements observations and interviews	High accuracy; Covers wide range of variables; Medium cost, Short implementation	Low precision for rare events; Trends not yet available	Guatemala Philippines
2. Other approaches
Compilations of forest management plans	Assembly of information obtained from obligatory management planning activities	Strong link to policy implementation	Scope normally limited to forest operations planning; May not cover all forests	Russia
Remote sensing based	Trends in land cover and land use from image classifications	Full cover information, Comparable time series	Limited scope of variables	Brazil
Independent reports over time	Not fully compatible surveys, often based on remote sensing, are compared with help of assumptions	Pragmatic (only) option in many countries	Relies on assumptions for comparisons	Ecuador
Case studies and/or Models	Conditions at selected sites are extrapolated	In-depth information for selected sites	Not represent-ative for entire country	Zambia
Expert estimates	Qualified guesses where only scant information exists	Quick and inexpensive	Unknown information quality	Sudan

4 Status of national forest monitoring

Two basic forest parameters were selected for this paper to illustrate the current status of national forest monitoring. These are highly relevant to the current discussions on deforestation and forest degradation in the climate change context:

1. Forest area change, i.e. the aggregated area change (net change) resulting from deforestation, afforestation and natural expansion of forests (FAO 2006c, p. 14)

2. Forest carbon stock changes (FAO 2006c, p. 31)

For each parameter, the monitoring approach was classified according to the categories in Table 3 for all countries and areas that reported to the Global Forest Resources Assessment 2005 (FRA 2005). It is assumed that essentially all available and relevant information have been used for the FRA 2005 reporting from countries (FAO 2006c). To make the results relevant in the climate change context, countries were grouped into those that are listed in the Annex I of the Kyoto Protocol, and those that are not. Broadly speaking this also divides the world into developed and developing countries.

229 countries and areas reported to FRA 2005. Of these, 41, mainly dependent territories, are not parties to the UNFCCC and are excluded in the presentations below; they represent only 0.5% of the global forest area. 148 countries of those reporting to FRA 2005 have signed the climate change convention, but are not “Annex 1” parties of the Kyoto Protocol and are therefore referred to as “non-Annex I countries”. 40 countries have signed both the convention and are listed in Annex I of the Kyoto Protocol and are referred to as “Annex I countries”.

The world’s forest area is nearly evenly distributed between Annex I countries (1.8 billion ha) and non-Annex I countries (2.1 billion ha). However, almost all current net loss of forest area occur in non-Annex I, i.e. developing, countries.

Figure 1 illustrates how the world’s countries monitor their forests with respect to the two selected parameters. Annex I countries have an even distribution of approaches across the monitoring categories, while the majority of non-Annex I countries (85% of countries in the case of forest area and 92% of countries in the case of carbon stock) rely on independent assessments, models or expert estimates rather than systematic inventories. Figure 2 illustrates examples of countries that have recently adopted systematic monitoring, but where trend estimates based on the systematic sample can not be determined until a second inventory round is completed. For non-Annex I countries, the monitoring of carbon stock is dominated by lack of data and expert estimates.

Table 4 provides the same breakdown, but shows the proportion of global forest area subject to the different monitoring approaches. The distribution is similar to that shown in Figure 1, implying that the lack of systematic monitoring approaches is spread among larger as well as smaller countries.

Table 4. Global forest area subject to monitoring approaches for the two parameters forest area change (ha/year) and forest carbon stock change (tonnes/ha/year)

	Forest area subject to monitoring approach, million hectares
	Parameter: Forest area change				Parameter: Forest carbon stock trend
Monitoring approach	non-Annex I	Annex I	Total	% of total	non-Annex I	Annex I	Total	% of total
National forest inventory	216	370	586	10	216	370	586	10
Forest management plans	49	859	908	2	16	853	869	1
Maps	129	164	292	6	0	0	0	0
Independent reports	883	85	968	42	744	65	810	35
Case studies / Models	622	11	633	29	30	30	60	1
Expert estimates	218	326	544	10	852	15	867	40
No data	0	0	0	0	258	482	741	12
Total	2,118	1,815	3,932	100	2,118	1,815	3,932	100

Figure 1. Forest monitoring approaches as applied by countries to estimate forest area change and forest carbon stock change. Based on FAO (2006c). Annex I indicates countries that are listed in Annex I of the Kyoto Protocol, Non-Annex I indicates countries that are not.

Figure 2. In centre, the World’s distribution of forests according to the FAO-led Global Forest Resources Assessment 2000. Clock-wise from top left, sample grids for national forest inventories in Guatemala, Lebanon, The Philippines and Cameroon as applied in FAO-supported national assessments between 2001 and 2005.

5 Discussion

Conclusions

The results show that forests are not systematically monitored in the majority of countries and especially not in non-Annex I countries on which current discussions on deforestation and forest degradation in the UNFCCC context focus. This means that accuracy and precision of reported information is not possible to determine for most countries. The quality problem of relying on independent assessments or expert estimates is well illustrated by Global Forest Resources Assessment 2000 findings, where the forest area change reported by African countries was overestimated by more than a factor two, compared with a systematic remote sensing survey (FAO 2001).

Conclusively, current monitoring of forests seems not sufficiently accurate or precise for an international protocol that would administer finances based on monitoring results of forest area or forest carbon storage.

In addition, at least one of the two parameters examined above, forest area, is very basic and usually the first to be considered in a national forest monitoring system. While carbon stock is a more novel forest parameter of interest, it is according to the applicable guidelines (IPCC 2003) simply a deterministic function of forest growing stock, another basic component of national forest inventories. Conclusively, there appear to be considerable opportunities for synergies between general requirements of national forest monitoring and the specific requirements of the climate change related arrangements.

Underlying causes

The call for better forest information for policy making is not new (e.g. Pinchot 1923). What may be surprising is that so astonishingly little has been done over the past decades to ensure the supply of solid forestry information – despite the considerable attention and focus on forestry in a large number of international fora as described above, and despite the considerable international development assistance provided to the forest sector (Holmgren & Persson 2002).

Before marching ahead and implementing national forest monitoring systems, however, there may be reason to investigate further why investments in such systems have not been sufficient in the past. As proposals for future studies, the following hypotheses are suggested:

- The politically driven demand for national forest and forestry information for policy-related decision making is smaller than the international dialogue on forests suggests. In competition with other Government activities, the monitoring of forests and forestry to determine actions that pay off in a relatively distant future are not prioritised;

- The awareness, knowledge, experience, engagement, opportunity or influence of forestry professionals and their institutions are insufficient to establish commitments to long-term national forest monitoring efforts;

- Methodologies to meet relatively new, and constantly changing, demands for information take time to develop, share and implement and may be outdated when finally delivered. Method developments may also have been biased by an over belief in the technical performance of remote sensing methods, which has slowed down the development of field-based monitoring.

Opportunities and challenges

To end this paper with some positive outlooks, there is reason to believe that awareness and prioritization of national forest monitoring systems are on the rise; examples include Brazil and Russia that have recently made decisions to invest in national forest inventories. Several developing countries work at institutional as well as field level to improve national forest monitoring in collaboration with FAO (FAO 2006d).

Integrated approaches to national forest monitoring promise to be effective. One reason is that many variables overlap between the policy issues to be addressed. As mentioned above, key variables for carbon monitoring are also key variables for monitoring productive functions of forests as well as biological diversity. Further, given that field sampling and inventory is the preferred approach to national forest monitoring, the benefits of the considerable effort of reaching field sample locations should be maximized by collecting many parameters while there. In conclusion, there appear to be strong arguments to incorporate currently debated requirements for forest carbon monitoring in the established general approach to national forest monitoring, thereby potentially improving the financial base for monitoring as well as enhancing future benefits of overall forest management.

There are further opportunities for integrating national monitoring approaches, extending beyond forests and forestry, to other land uses and related natural environments. In many countries, the forest resources extend over all land, including woodlands and trees outside of forests. As these resources are important to forest-sector policies, national forest monitoring often extend to all land. Sampling all land provides opportunities for development and implementation of inter-sectoral national land-use monitoring as the additional cost to collect, e.g., agriculture variables may be small. Such inter-sectoral approaches may (a) help enhance institutional collaboration in countries, (b) make better use of scarce inventory resources and (c) provide improved possibilities for inter-sectoral land use analyses and policy formulation (FAO 2006d).

Providing the relevant information to the relevant people at the relevant point in time at relevant cost are fundamental challenges for national forest monitoring systems. Competition for public finances and the fact that returns on the monitoring investment accrue in a relatively distant future add to the difficulties. Yet, our livelihoods depend heavily on the future of forest resources and the environment, calling for wise decisions on all levels, which suggests that the market for monitoring and quality information could expand.

6 References

Andersson, C. 2006. Data collection through interviews. In: National Forest Assessment Knowledge Reference, Food and Agriculture Organization, Rome. www.fao.org/forestry/fra-knowledgeref (accessed 3 December 2006)

CPF (Collaborative Partnership on Forests). 2006. Streamlining forest-related reporting – Joint information framework. www.fao.org/forestry/cpf-mar (accessed 3 December 2006)

Davenport, D. 2005. An alternative explanation for the failure of the UNCED forest negotiations. Global Environmental Politics 5(1):105-130.

FAO. 1998. Forestry policies in the Caribbean. FAO Forestry Paper 137.

FAO. 2001. Global Forest Resources Assessment 2000 – main report. FAO Forestry Paper 140. www.fao.org/forestry/fra

FAO. 2005. State of the World’s Forests 2005. Food and Agriculture Organization, Rome. www.fao.org/forestry/sofo

FAO. 2006a. FAO Forestry Country Profiles. Food and Agriculture Organization, Rome. www.fao.org/forestry/country-info (accessed 3 December 2006)

FAO. 2006b. What is sustainable forest management? Food and Agriculture Organization, Rome. www.fao.org/forestry/site/24447/en (accessed 3 December 2006)

FAO. 2006c. Global Forest Resources Assessment 2005 – Progress towards sustainable forest management. FAO Forestry Paper 147. www.fao.org/forestry/fra2005

FAO. 2006d. Support to national forest assessments. Food and Agriculture Organization, Rome. www.fao.org/forestry/site/24673/en (accessed 3 December 2006)

IPCC. 2003. Good practice guidance for land-use change and forestry. IGES. Japan.

Gillis, M.D., Omule, A.Y, & Brierley. T. 2005. Monitoring Canada’s forests: the National Forest Inventory. Forestry Chronicle 81(2):214-221.

Holmgren, P. & Persson, R. 2002. Evolution and prospects of global forest assessments. Unasylva Vol. 53, No. 210:3-9 www.fao.org/forestry/unasylva

Holmgren, P. & Thuresson, T. 1998. Satellite remote sensing for forestry planning - a review. Scandinavian Journal of Forest Research 13(1):90-110.

Kleinn, C., Ramirez, C., Holmgren, P., Valverde, S.L. & Chavez. G. 2005. A national forest resources assessment for Costa Rica based on low intensity sampling. Forest Ecology and Management 210:9-23.

Kleinn, C. 2006. Observation and Measurement. In: National Forest Assessment Knowledge Reference, Food and Agriculture Organization, Rome. www.fao.org/forestry/fra-knowledgeref (accessed 3 December 2006)

Persson, R. 2005. Where is the United Nations Forum on Forests going? International Forestry Review 7(4):348-357

Pinchot, G. 1923. Foreword. In: Zon, R. & Sparhawk, W. Forest resources of the world, Volume I. McGraw-Hill New York.

Ranneby, B., Cruse, T., Hägglund, B., Jonasson, H. & Swärd, J. 1987. Designing a new national forest survey for Sweden. Studia Forestalia Suecica 177.

Ruis, B. 2001. No forest convention but ten tree treaties. Unasylva 206:3-13. www.fao.org/forestry/unasylva