Climate change and forests

[노남진]

<원문출처 www.fao.org/docrep/003/y0900e/y0900e06.htm >

Climate change and forests

The mitigation of global climate change through forestry was first proposed in the 1970s (Dyson, 1977). It was not until the late 1990s, however, that international negotiations considered this potential at a global level, calling for a definition and quantification of the role of forests and proposing a mechanism for international collaboration.

In 1992, the Framework Convention on Climate Change (FCCC) was adopted as a consequence of worldwide concern about global warming. The Convention aims at stabilizing the concentration of greenhouse gases in the atmosphere in an effort to reduce human-induced disturbances to the global climate system. The industrialized countries and countries in transition that are parties to the FCCC (Annex 1 Parties) committed themselves to carrying out national inventories of greenhouse gas emissions and carbon sinks, and to working towards voluntary goals in the reduction of emissions. At the third meeting of the Conference of the Parties, held in Kyoto, Japan in December 1997, an additional legally binding instrument was adopted: the Kyoto Protocol to the Framework Convention on Climate Change. Thirty-nine developed countries (comprising a slightly modified list of Annex 1 Parties) committed themselves to reducing their greenhouse gas emissions between 2008 and 2012 by at least 5 percent compared with 1990 levels. Parties can meet this commitment by reducing sources or protecting or enhancing sinks of greenhouse gases. The Kyoto Protocol foresees the inclusion of changes resulting from direct human-induced land use change and forest activities, limited to afforestation, reforestation and avoidance of deforestation.

The Kyoto Protocol also sets up a framework for the transfer of emission credits between parties. Three flexible mechanisms were introduced that permit signatory countries to meet their commitments partially or fully: projects undertaken jointly by Annex I Parties (Joint Implementation), projects between Annex I and non-Annex I Parties (Clean Development Mechanism) and emissions trading. Although the Kyoto Protocol has not entered into force and it is as yet¹⁴ undecided whether forests will be included as sinks within the ambit of the flexible mechanisms, the role of forests in the context of climate change merits a close look because of the impact that the outcome of related decisions could have.

GLOBAL CARBON CYCLE

The International Panel on Climate Change (IPCC)¹⁵ estimates that the global mean temperature of the earth's surface has increased by 0.3 to 0.6ºC over the past 100 years (IPCC, 2000). Predictions are that global warming will cause significant variations in climatic patterns over the next century that may have negative impacts on regional and global biomes. It is now generally accepted that this change in global temperature is caused primarily by rising atmospheric concentrations of greenhouse gases, principally carbon dioxide (CO2), methane (CH4) and nitrous oxide (N2O). The most important of these greenhouse gases, CO2, accounts for some 65 percent of the "greenhouse effect". The rise of atmospheric CO2 concentrations since the beginning of the industrial revolution has been caused by anthropogenic activity, in particular the combustion of fossil fuels, cement manufacture and deforestation.

Terrestrial ecosystems play a significant role in the global carbon cycle. An estimated 125 gigatonnes (Gt)¹⁶ of carbon are exchanged annually between vegetation, soils and the atmosphere, accounting for two-fifths of the total exchange of carbon between the earth and the atmosphere (see Figure 18). Forests account for some 80 percent of this exchange. While the world's forests are absorbing carbon, they are also releasing it. Deforestation is a significant source of carbon emissions; evidence suggests that deforestation in the 1980s may have accounted for one-fourth of all anthropogenic carbon emissions (Houghton, 1999).¹⁷ However, it has been suggested that the terrestrial biosphere could be managed over the next 50 years to conserve or sequester 60 to 87 Gt of carbon in forests and another 23 to 44 Gt of carbon in agricultural soils (Brown et al., 1996).

FIGURE 18

An estimate of the current global carbon cycle¹

¹All numbers are in gigatonnes (Gt) of carbon (1 Gt = 1 billion tonnes).

Note: The magnitude of the fluxes between the atmosphere and the oceans and terrestrial biosphere is still uncertain and is the subject of ongoing research.

THE ROLE OF FORESTS IN THE GLOBAL CARBON BUDGET

Carbon stocks in forest ecosystems

Carbon accumulates in forest ecosystems through the absorption of atmospheric CO2 and its assimilation into biomass. Carbon is stored in living biomass, including standing timber, branches, foliage and roots; and in dead biomass, including litter, woody debris, soil organic matter and forest products. Any activity that affects the amount of biomass in vegetation and soil has potential to sequester carbon from, or release carbon into, the atmosphere.

Overall, forests contain just over half of the carbon residing in terrestrial vegetation and soil, amounting to some 1 200 Gt of carbon (see Figure 19).

FIGURE 19

Terrestrial carbon stocks by ecosystem

Note: 1 gigatonne (Gt) = 1 billion tonnes.
Source: Dixon et al., 1994; Schlesinger, 1997.
Boreal forests account for more carbon than any other terrestrial ecosystem (26 percent of total terrestrial carbon stocks), while tropical and temperate forests account for 20 and 7 percent, respectively (Dixon et al., 1994). In comparison with other vegetation in other terrestrial ecosystems, forest vegetation has a very high carbon density (see Figure 20).

FIGURE 20

Above-ground carbon density for selected vegetation types

Source: IPCC, 2000.

The carbon stored in the soil and litter of forest ecosystems also makes up a significant proportion of the total carbon pool. Globally, soil carbon represents more than half of the stock of carbon in forests. There are, however, considerable variations among ecosystem and forest types. Between 80 and 90 percent of the carbon in boreal ecosystems is stored in the form of soil organic matter, whereas in tropical forests the carbon is fairly equally distributed between vegetation and soil (see Table 10). The primary reason for this difference is the influence of temperature on the relative rates of production and decay of organic matter. At high latitudes (i.e. in cooler climates), soil organic matter accumulates because it is produced faster than it can be decomposed. At low latitudes, however, warmer temperatures encourage the rapid decomposition of soil organic matter and subsequent recycling of nutrients.

TABLE 10 Carbon density and stock of vegetation and soils for different ecosystems
Ecosystem	Country/region	Vegetation carbon density (tonnes/ha)	Soil carbon density (tonnes/ha)	Vegetation carbon stock (Gt)	Soil carbon stock (Gt)	Total carbon stock (Gt)
Boreal	Russian Federation	83	281	74	249	323
	Canada	28	484	12	211	223
	Alaska	39	212	2	11	13
Temperate	United States	62	108	15	26	41
	Europe	32	90	9	25	34
	China	114	136	17	16	33
	Australia	45	83	18	33	51
Tropical	Asia	132-174	139	41-54	43	84-97
	Africa	99	120	52	63	115
	Americas	130	120	119	110	229
Note: 1 gigatonne (Gt) = 1 billion tonnes. Source: Dixon et al., 1994.

Carbon fluxes from forest ecosystems

All forest biomes have undergone major changes in distribution since the height of the last ice age (18 000 years ago), when the climate was both cooler and more arid than it is today. Boreal and northern temperate forests were squeezed between advancing ice sheets and steppe tundra from the north and expanding semi-desert and steppe tundra from the south, while tropical rain forests retreated into small pockets as savannah expanded. The amount of carbon stored in terrestrial biomes was 25 to 50 percent lower than at present. Terrestrial carbon storage peaked in the warm, moist early Holocene period about 10 000 years ago and subsequently declined by about 200 Gt to reach today's level (2 200 Gt of carbon), probably because of a gradual cooling and aridification of the climate.

Prior to the nineteenth century, humans exerted only a modest influence on terrestrial carbon storage through fire, fuel use and deforestation, but since the outset of the industrial revolution, human activities have had a major effect on the global carbon cycle. Between 1850 and 1980, more than 100 Gt of carbon were released into the atmosphere as a result of land use changes, representing about one-third of the total anthropogenic carbon emissions over this period (Houghton, 1996).

Until the late nineteenth century, most forest clearing and degradation took place in temperate regions. In the twentieth century, the area of temperate forest largely stabilized and tropical forests became the primary source of carbon emissions from terrestrial ecosystems (Houghton, 1996). Today, forest cover in developed countries is increasing slightly: between 1980 and 1995 there was an average increase of 1.3 million ha per year (FAO, 1999d). In recent decades, many temperate forest regions (such as Europe and eastern North America) have become moderate carbon sinks through the establishment of plantations, the regrowth of forests on abandoned agricultural lands, and increased growing stock in forests.
In contrast, tropical forests have become a major source of carbon emissions; the rate of tropical deforestation is estimated to have been 15.5 million ha per year in the period 1980-1995 (FAO, 1999d).

Net carbon emissions resulting from land use change in the 1980s are estimated to be between 2 and 2.4 Gt per year (see Figure 21), equivalent to between 23 and 27 percent of all anthropogenic emissions (Houghton, 1999; Fearnside, 2000). Tropical deforestation accounts for most of the carbon emissions from land use change.

FIGURE 21

Carbon emissions from land use change

Source: Houghton, 1999.

The burning of biomass also releases other greenhouse gases, including methane and nitrous oxide. Burning of forest biomass accounts for 10 percent of global methane emissions. Forest degradation also results in carbon loss. An estimated net annual emission of 0.6 Gt of carbon was due to the degradation of forests in the tropics during the 1980s (Houghton, 1996). In tropical Asia, the loss of carbon resulting from forest degradation almost equals that caused by deforestation.

There is accumulating evidence that human-induced changes in concentrations of atmospheric gases are affecting the carbon cycle in forests. Global atmospheric CO2 concentrations have risen from 280 ppm before the industrial revolution to 370 ppm in 2000, and nitrogen deposition rates in forests near industrial regions have increased substantially. Both these effects are likely to lead to an increase in plant growth and productivity. Permanent forest sample plots in climax forests of North and South America have shown significant increases in forest biomass in recent years. Other evidence for enhanced carbon uptake in forest regions comes from micrometeorological measurements of CO2 fluxes above forests and assessments of atmospheric CO2 distributions at continental scales. Studies suggest that, through the combined effects of reforestation, regrowth of degraded forests and enhanced growth of existing forests, between 1 and 3 Gt of carbon are absorbed per year, approximately offsetting the global emissions from deforestation (Malhi, Baldocchi and Jarvis, 1999).

CLIMATE CHANGE AND FORESTS¹⁸

If the temperature at the earth's surface increases during the twenty-first century as predicted, all ecosystems will experience the most rapid period of climate change since the end of the last ice age. The distribution and composition of forests will be affected by this change, and management strategies will need to accommodate the prospect of rapidly shifting climate zones and ecosystem margins.

Box 16 presents the predicted impacts on major forest types under climate change scenarios as indicated by IPCC models of global climate change in the twenty-first century. The models show a fair degree of consistency in their predictions of global warming, with less agreement on changes in precipitation. All these model scenarios assume that no big "surprises" will occur.¹⁹ Using IPCC climate prediction scenarios, the key changes expected towards the end of the twenty-first century are:

• the atmospheric concentration of CO2 will approximately double;

• the mean global temperature will increase by 1.5 to 4.5°C;

• precipitation will increase globally by 3 to 5 percent;

• the sea level will rise by about 45 cm.

Regional climate predictions are needed to determine the impacts on forests. There is a fair degree of confidence in most regional temperature predictions. The largest increases in temperature will be in the northern high latitudes, with lower increases nearer the tropics and in regions with a strong oceanic influence. Although precipitation will increase globally, regional predictions are less reliable. Overall, the key climatic changes controlling forest growth responses will be temperature increases at high latitudes and changes in rainfall at low latitudes. Any regions with increased temperature and unchanged or reduced rainfall will experience significant reductions in soil moisture, which will constrain plant growth and increase the likelihood of fire. Large outbreaks of fire may lead to significant losses of forest cover.

Existing forest stands may persist for some time under a changed climate, but long-term responses to climate change will depend on the capability of species to adapt to the new conditions or to change their geographic distributions. This capability will be determined by the variation within and between species in their physiological responses to changes in temperature, CO2 concentration, soil moisture and, in some areas, increased nitrogen deposition. It will also depend on soil types and the ecological relationships between species that affect pollination, dispersal and damage through herbivory or pest and pathogen attacks. The nature of the landscape and the intensity of human activities will also be determining factors. For example, habitat fragmentation will affect how effectively species can change their geographic range in response to ecosystem shifts. Mountains may be particularly important refugia in a warming climate because many species will find it easier to shift their range upwards in altitude, to a cooler climate, than upwards in latitude over large distances. Changes in species distribution may lead to new species assemblages and may involve species losses.

BOX 16

Impacts of climate change on different forest types

Boreal forests will experience the largest temperature increases of all forests. The warming effect is expected to be greater in winter (4°C above the levels of the 1970s by the middle of the twenty-first century) and slightly lower in summer (2.5 to 3°C above the levels of the 1970s). Reduced moisture in the soil during summer will increase drought stress and the frequency and extent of wildfires. Climate zones are expected to shift northwards by as much as 5 km per year. Boreal forests will make gains in areas to the north, but will experience dieback and replacement at their southerly extremes. Changes in the frequency, intensity and extent of wildfires in response to increased heat stress will play a critical role in determining the dynamics of the changes at the southern fringe of the boreal forests. Models used to predict the long-term potential changes in the distribution of vegetation suggest that the overall response may be either a reduction (by up to 36 percent) or an expansion (by up to 16 percent) in boreal forest area, although a reduction is more likely. Few tree species are likely to become extinct, but local species loss may be significant.

Temperate forests will be most affected by climate warming at higher latitudes (2.6°C above the levels of the 1970s by the middle of the twenty-first century) and by changes in rainfall at lower latitudes. Drought stress at certain low-latitude margins (such as the Mediterranean and southwestern United States) may lead to significant dieback, while increased temperatures may enhance growth at higher latitudes. Climate zones will shift towards the poles at rates of up to 5 km per year. The potential area available for temperate forest growth is likely to expand by between 7 and 58 percent. The high level of fragmentation of many temperate forests is likely to limit effective dispersal of some tree species (and have an impact on forest-based wildlife). This may lead to significant species losses locally.

Tropical forests are expected to warm by 2°C above the levels of the 1970s by the middle of the twenty-first century, with larger increases in continental interiors. Changes in rainfall regime, however, are likely to be more important than changes in temperature, although model predictions of regional rainfall patterns vary substantially. Where there are reductions in rainfall and higher temperatures, reduced soil moisture is expected to be the most significant threat to tropical forests. These effects may increase vulnerability to fire or lead to significant dieback or changes in vegetation types in marginal areas. Interannual variability as a result of large-scale climatic events (such as those caused by the El Niño phenomenon) may exacerbate rainfall extremes. Depending on future climate scenarios, the potential tropical forest area could shrink by as much as 30 percent or expand by up to 38 percent. In most tropical regions, however, the impact of human activities such as deforestation or burning will be more important than climate change in determining forest cover. A shrinking of the area of tropical forests, particularly of moist tropical forests, would be likely to result in significant species losses.

Tropical montane cloud forests are expected to warm by to 2°C by the middle of the twenty-first century, but they are most threatened by changes in the height of the cloud base, on which they depend for dry season water supply. Cloud base heights are likely to rise by as much as 2 m per year - which would affect the species in these forests. Where mountains are isolated and insufficiently high to accommodate upward changes in cloud height, climate change may lead to the local, if not total, extinction of some montane vegetation species (many of which are endemics). There is evidence from cloud forest in Monteverde, Costa Rica that such changes are already occurring. Cloud forests may be harbingers of climate change effects on global forest ecosystems.

Mangrove forests are expected to be able to adapt to rising temperatures but may be threatened by rising sea levels. This threat will be particularly acute for sediment-poor coasts, such as those found on small islands, and in areas where inland dispersal of forest species is constrained by human land use.

Source: WCMC, in press.

Changes in forest cover could induce feedback effects on the climate by modifying surface temperatures and by influencing atmospheric CO2 concentrations. Forests have a lower albedo (i.e. they reflect less light) than other ecosystems and, through their extensive root systems, have more access to soil water than other types of vegetation. In consequence, they absorb more solar energy, which can lead to heating, and lose more water through evaporation, which can lead to cooling. In tropical zones, evaporation processes tend to dominate and the net effect of forests is to cool and moisten the atmosphere. At higher latitudes, albedo effects are more important, thereby leading to local warming.

TABLE 11 Overview of terrestrial carbon management strategies and potential land use and forest activities
Carbon management strategy	Type of land use and forest activity
Carbon sequestration	Afforestation, reforestation and restoration of degraded lands Improved silvicultural techniques to increase growth rates Implementation of agroforestry practices on agricultural lands
Carbon conservation	Conservation of biomass and soil carbon in existing forests Improved harvesting practices (e.g. reduced impact logging) Improved efficiency of wood processing Fire protection and more effective use of burning in both forest and agricultural systems
Carbon substitution	Increased conversion of forest biomass into durable wood products for use in place of energy-intensive materials Increased use of biofuels (e.g. introduction of bioenergy plantations) Enhanced utilization of harvesting waste as feedstock (e.g. sawdust) for biofuel
Source: Bass et al., 2000.

CARBON MANAGEMENT STRATEGIES

There are three possible strategies for the management of forest carbon (see Table 11). The first is to increase the amount or rate of carbon accumulation by creating or enhancing carbon sinks (carbon sequestration). The second is to prevent or reduce the rate of release of carbon already fixed in existing carbon sinks (carbon conservation). The third strategy is to reduce the demand for fossil fuels by increasing the use of wood, either for durable wood products

(i.e. substitution of energy-intensive materials such as steel and concrete) or for biofuel (carbon substitution). These strategies are not mutually exclusive. A number of carbon sequestration and carbon conservation initiatives have already been developed, including Activities Implemented Jointly (AIJ)²⁰ under the FCCC and Land Use Change and Forestry (LUCF) carbon projects.

Carbon sequestration

The carbon sequestration potential of afforestation/reforestation is specific to the species, site and management involved, and it is therefore very variable. Typical sequestration rates for afforestation/reforestation, in tonnes of carbon per hectare per year, are: 0.8 to 2.4 tonnes in boreal forests, 0.7 to 7.5 tonnes in temperate regions and 3.2 to 10 tonnes in the tropics (Brown et al., 1996). The sequestration potential for agroforestry practices is even more variable, depending on the planting density and production objectives of the system.

Assuming a global land availability of 345 million ha for afforestation/reforestation and agroforestry activities, Brown et al. (1996) estimate that approximately 38 Gt of carbon could be sequestered over the next 50 years - i.e. 30.6 Gt by afforestation/reforestation and 7 Gt through the increased adoption of agroforestry practices (see Figure 22). Studies of tropical regions indicate that an additional 11.5 to 28.7 Gt of carbon may be sequestered through the assisted regeneration of about 217 million ha of degraded land.

FIGURE 22

Potential contribution of afforestation/reforestation and
agroforestry activities to global carbon sequestration, 1995-2050

Source: Data from Brown et al., 1996.

However, the actual availability of land for forest activities may be considerably less when full account is taken of social and economic factors. Only one-third of ecologically suitable land may actually be available for afforestation/reforestation activities (Houghton, Unruh and Lefebvre, 1991). Under this scenario, afforestation/reforestation and agroforestry activities would absorb about 0.25 Gt and the restoration of degraded lands a further 0.13 Gt of carbon per year.

Silvicultural activities that increase the productivity of forest ecosystems, such as timely thinning, can increase forest carbon stocks to some extent. However, compared with afforestation/reforestation, the effect of varying silvicultural systems on total carbon stocks is relatively low (Dixon et al., 1993).

Carbon conservation

While the most effective means to reduce atmospheric concentrations of CO2 is the reduction of emissions from fossil fuel combustion, in terms of land use change and forestry the conservation of existing forest carbon stocks has technically the greatest potential for rapid mitigation of climate change. As the majority of carbon emissions from deforestation occur within a few years of forest clearance, reducing the rate of deforestation will produce a more immediate effect on global atmospheric CO2 levels than will afforestation/reforestation measures, in which similar volumes of carbon may be removed from the atmosphere but over a much longer period.

The potential for carbon conservation through the maintenance of forest cover is dependent on the assumed baseline for non-project deforestation (i.e. "business as usual"). In principle, 1.2 to 2.2 Gt of carbon could be conserved annually if deforestation were stopped completely (Dixon et al., 1993). However, while carbon revenues could improve the economics of forest land, projects will also have to address the underlying causes of deforestation and unsustainable forest use to achieve effective carbon conservation. Brown et al. (1996) estimate that a reduction in deforestation in tropical regions could feasibly conserve 10 to 20 Gt of carbon by 2050 (0.2 to 0.4 Gt per year).

The conservation of carbon stocks in existing forest may be achieved through improved management practices. Potentially, the most important is the use of reduced impact logging (RIL) in the tropics. Conventional logging practices may result in a high level of damage to the residual stand, with up to 50 percent of remaining trees damaged or killed (Kurpick, Kurpick and Huth, 1997). The application of RIL techniques can reduce the level of damage to the residual stand by 50 percent (Sist et al., 1998) and hence reduce the level of carbon emissions associated with logging. Nabuurs and Mohren (1993) calculated that long-term carbon conservation resulting from RIL in tropical rain forest may be between 73 and 97 tonnes per hectare. Given that an estimated 15 million ha of tropical forest is logged each year (Singh, 1993), the majority of which is considered to be unsustainable (Poore et al., 1989), the potential for increased carbon storage is large. The additionality of carbon conserved through RIL techniques is dependent on the assumption that conventional logging would continue in the absence of intervention, and there is concern about how to quantify the changes in carbon stocks associated with changes in harvesting practices (IPCC, 2000, Chap. 4).

Wildfires result in large losses of carbon from forests every year. Weather conditions brought on by climate change, such as the enhanced El Niño phenomenon, increase the potential risk of wildfires. Fire management practices have the potential to conserve carbon stocks in forests. However, if fire management is to be effective, fire prevention and firefighting efforts must be combined with land use policy changes and measures to address the needs of rural populations. There could also be problems in assessing the baseline for fire prevention projects, which will be dependent on interactions between human factors and stochastic factors such as weather.

Carbon substitution

Biofuels currently provide 14 percent of the global primary energy supply. In developing countries, biofuels account for one-third of the total energy supply. If current biofuel use were to be replaced by fossil fuel-derived energy, an additional 1.1 Gt of carbon per year would be released into the atmosphere (IPCC, 2000, Chap. 5). In contrast to the combustion of fossil fuel, the use of sustainably produced biofuels does not result in a net release of CO2 into the atmosphere, since the CO2 released through the combustion of biofuels is taken up by regrowing biomass. The substitution of fossil fuels by sustainable biofuels will therefore result in a reduction of CO2 emissions that is directly proportional to the volume of fossil fuel replaced. Predictions of the future contribution of biofuels in meeting energy requirements range from 59 to 145 x 1018 J for 2025 and 94 to 280 x 1018 J for 2050 (Bass et al., 2000). The future usage will depend to a large extent on the development of technologies that permit an efficient use of biofuels, such as the gasification of wood products.

New biofuel plantations will also have a long-term positive sequestration effect if they replace a land use with a lower sequestration rate. Although the long-term average carbon density of a forest managed for biofuels (particularly for short-rotation coppice) will be lower than an unharvested forest or long-rotation plantation, this forest use stores more carbon than most non-forest land uses. Conversely, if natural forests are replaced with short-rotation coppice for biofuel production, the beneficial effect of fossil fuel substitution will be lost because of the emissions resulting from forest conversion.

The use of wood products in place of materials that are associated with the release of large volumes of carbon dioxide (either during processing - such as cement - or through energy consumption - such as steel) could also lead to significant net reductions in CO2 emissions.

Experiences in Land Use Change and Forestry (LUCF) project-based activities

There are currently 16 approved international AIJ projects involving Land Use Change and Forestry (FCCC, 2000). Table 12 provides a summary of a representative set of LUCF projects currently under implementation, covering about 3.5 million ha (IPCC, 2000, Chap. 5). Eighty-three percent of this area is managed for the conservation of carbon in existing forests, either through forest protection (zero harvesting) or forest management (sustained production). Long-term carbon conservation by these projects varies from 40 to 108 tonnes per hectare from forest management and from 4 to 252 tonnes per hectare from forest protection. The estimated total lifetime sequestration effect of these projects is 5.7 million tonnes of carbon from forest management and 40 to 108 million tonnes from forest protection.
A further 180 000 ha are managed for afforestation/reforestation activities and will offset²¹ an estimated 21.7 million tonnes of carbon over project time scales. Two projects, covering 200 000 ha, involve agroforestry and are expected to offset an additional 10.8 million tonnes of carbon.

The cost per tonne of carbon for the projects described in Table 12 ranges from US$0.1 to US$15. However, it should be noted that the approaches for calculating the costs of carbon sequestration vary considerably between projects, and long-term estimates may need to be revised upwards. The eventual uptake of carbon sequestration potential will be dependent on the comparative costs of emissions reduction from the energy sector; some studies indicate that the market for carbon from the forest sector may be below 1 Gt.

TABLE 12 Comparison of selected LUCF projects
Project type	Number of projects	Area (million ha)	Carbon stored (million tonnes)	Carbon stored (tonnes/ha)	Costs (US$/tonne of carbon)
Sequestration
Afforestation/reforestation	8	0.18	21.7	26-328	1-28
Agroforestry	2	0.20	10.8	56-165	0.2-10
Conservation
Protection	7	2.90	40-108	4-252	0.1-15
Management	4	0.33	5.7	0.2-85	0.3-8
Source: IPCC, 2000, Chap. 5.

Besides the question of how to calculate the costs of carbon sequestration, an important issue is the methodology for carrying out carbon accounting. Box 17 discusses carbon accounting at both the national and project levels.

BOX 17

Accounting for carbon sequestration by forestry

The accounting of greenhouse gas emissions attributable to nations, companies and industrial processes has become an important component of international agreements and national policies to address climate change.

The accounting of carbon benefits attributable to forest activities is of significant interest because of the forest sector's potential to contribute to the achievement of national emissions reduction targets negotiated under the FCCC, and also because of the potential value of forestry projects in offsetting emissions from specific businesses or business activities.

NATIONAL CARBON ACCOUNTING

National emissions or uptake of carbon by forests are accounted on an annual basis and are expressed in tonnes of CO2 released or sequestered. Advances towards the targets set under the Kyoto Protocol are measured in terms of emissions or uptake relative to 1990. Under Article 3.3 of the Protocol, only the uptake of carbon by afforestation, reforestation and avoided deforestation may be counted towards national emissions reduction targets. The precise definitions of afforestation, reforestation and avoided deforestation are still being discussed.

PROJECT CARBON ACCOUNTING

Additionality and baselines

While national emissions and uptake of carbon are measured in absolute terms within national boundaries, the effect of forestry projects is measured relative to a hypothetical "without project scenario" or "baseline". The definition of a project baseline could be derived in a number of ways, including the extrapolation of previous trends in land use change, the expected impacts of current standard forestry practices, or by modelling of the social and economic pressures on forest resources. Standard methods have yet to be agreed on. When a project and a baseline case are compared, so-called "additionality" tests may be applied to ascertain whether carbon sequestration is attributable to the project or simply to incidental factors, including shifts in policy or socio-economic conditions outside the scope of the project.

Project boundaries and leakage

The setting of project boundaries will have an important effect on the emissions reductions attributed to project activities. If a project envisages the protection of a particular area of forest but involves the shifting of forest clearing to another area, there is potential for a "leakage" of project benefits. Similarly, if an afforestation project leads to a reduction in timber prices and subsequent reduced investment in commercial plantations or increased clearance of forest land to fulfil subsistence food requirements, the net sequestration will be reduced. Project boundaries also need to be set to include all flows or stocks of carbon that might be significantly affected by project activities; this may include carbon stored in harvested timber products.

Project time scales and crediting

The long time scales associated with forest growth, particularly in temperate and boreal regions, and the potential reversibility of carbon gains through forest activities are key features of Land Use Change and Forestry (LUCF) projects. A number of alternative conventions for crediting the carbon sequestration or avoided emissions from forestry have been proposed:

• Ex ante, or upfront, crediting of future carbon sequestration, which would enable project developers to take credit for carbon uptake and storage that will occur in the future. This would make project development relatively easy but would require other mechanisms to guarantee fulfilment and long-term maintenance of carbon gains.

• Staged crediting, in which credit for carbon sequestration would be accrued in stages, so that project developers would have to demonstrate carbon gains before gaining recognition.

• Ex post, or delayed, crediting, in which credit for sequestration would only be given after carbon had been stored for a certain time, for example 40 or 50 years. This type of crediting would provide a strong measure of guarantee regarding the effectiveness of carbon offset projects but would provide little incentive for their development.

Box 18 gives some examples of activities undertaken in LUCF projects.

BOX 18

Examples of Joint Implementation (JI) projects currently in operation

RIO BRAVO CONSERVATION AND MANAGEMENT AREA, BELIZE

The Rio Bravo project involves the protection of 14 000 ha of "endangered" forest land and the development of a sustainable management programme for an additional 46 000 ha of forest. The project is managed by a Belizean NGO, called the Programme for Belize, and is financed in part by carbon offsets sold to a group of United States electricity utilities. In total, an estimated 2.5 million tonnes of carbon would be conserved over the 40-year life of the project, with an average potential of 36 tonnes of carbon per hectare at a cost of US$3 per tonne of carbon. The baseline case against which carbon benefits are calculated assumes that, without the project, the whole area of endangered forest would be deforested within five years. The land was previously privately owned and would probably have been sold to neighbouring farmers who had expressed interest in expanding their farms.

REDUCED IMPACT LOGGING (RIL) IN SABAH, MALAYSIA

Under this project, involving the forest concession Innoprise Corporation Sdn. Bhd. and New England Power of the United States, RIL techniques were adopted for use in 1 400 ha of dipterocarp forest in Malaysia for a period of two years. The resulting 50 percent reduction in damage to the forest vegetation (compared with conventional logging methods) conserved an estimated 40 tonnes of carbon per hectare at a cost of about US$8 per tonne. The calculation of carbon benefits assumes that the use of conventional logging methods would have continued if the project had not intervened. Thus, the carbon conserved is additional for only as long as conventional logging practices would have continued.

SCOLEL TE COMMUNITY FORESTRY PROJECT IN MEXICO

This project was set up by the University of Edinburgh and the Edinburgh Centre for Carbon Management in the United Kingdom and El Colegio de la Frontera Sur in Mexico, with funds from the United Kingdom's Department for International Development (DFID). The aim is to develop model planning and administrative systems by which small farmers can gain access to carbon markets. Under the project, small farmers and local communities identify reforestation, agroforestry and forest restoration activities that are both financially beneficial and intended to sequester or conserve carbon. The proposed activities are entered into a planning and eval‎uation system and the offsets are sold through a trust fund managed by a local NGO, Ambio. The systems are now well developed and carbon has been sold to various purchasers, including the International Automobile Federation. Around 300 farmers, with an average of about 1 ha of forest each, are currently involved. The average carbon sequestration potential is 26 tonnes per hectare at a cost of US$12 per tonne. The system applies a simple additionality criterion: carbon sequestration is deemed to be additional if one of the objectives of the planned afforestation is carbon sequestration. The baseline used is the mean carbon storage of the previous land use, the assumption being that the land use would have continued in the absence of project intervention.

CARBON SINK PROJECT IN MATO-GROSSO, BRAZIL

The project aims to sequester carbon in a plantation of native species established on degraded land. The plantation covers about 5 000 ha and is located on a private landholding of 15 000 ha. Only the plantation constitutes the carbon sink, but efforts are also being made to conserve the natural forest, even though it does not count towards the carbon balance. The project is funded by the French car manufacturer, Peugeot, as part of its environmental preservation policy. It is managed by ONF Brazil, a subsidiary of the Office National des Forêts of France, and the Instituto Pro Natura, a Brazilian NGO. The duration of the project is 40 years. Peugeot's contribution is US$10 million. The objective is to maximize carbon sequestration, while using local species and maintaining or enhancing biological diversity in the area. The project represents a first step towards re-establishing natural forest through the rehabilitation of pastures and the elimination of introduced grass species. The baseline is based on the continuation of the previous land use.

MANAGEMENT OF RESERVA FORESTAL MALLECO, CHILE

The project purchases the right to implement a management plan on 16 625 ha of natural, state-owned forest. Chilean regulations prohibit the transformation of this kind of forest to exotic species plantations. The purpose of the project is to promote sustainable management of the forest to prove its feasibility both in economic and carbon sequestration terms. The project is managed by the Corporación Nacional Forestal of Chile (CONAF) and the Office National des Forêts of France and it is sponsored by the French Fund for World Environment. The project involves adjustment of the methods of measuring carbon flux for this kind of ecosystem. The calculation of carbon benefits will be done from a "without project scenario" baseline.

Sources: Stuart and Costa, 1998; IPCC, 2000, Chap. 5; Tipper et al., in press; and Conseil Général du Genie Rural des Eaux et des Forêts, personal communication.

CONCLUSIONS AND FUTURE ISSUES

Forests are an important component of the global carbon cycle. They both influence and are influenced by climate change, and their management or destruction will have a significant impact on the course of global warming in the twenty-first century.

Forest ecosystems contain more than half of all terrestrial carbon. They account for about 80 percent of the exchange of carbon between terrestrial ecosystems and the atmosphere. Although forest ecosystems absorb between 1 and 3 Gt of carbon annually through regrowth in degraded forests, reforestation, and CO2 and nitrogen fertilization effects, they release about the same amount (2 Gt) each year through deforestation. Deforestation in the 1980s may have accounted for one-quarter of all anthropogenic carbon emissions.

If predicted climate changes materialize, the impacts on forests are likely to be dramatic and long-lasting. Forest ecosystems may persist for some time under changed climatic conditions, but long-term responses will depend on the capability of species to continue to adapt to the new conditions or to change their geographic distribution.

Forest management can contribute towards emissions reductions and to carbon sequestration (see Figure 23). The conservation of existing carbon stocks in forests is potentially a more powerful tool than carbon sequestration. However, forestry measures alone will not be enough to halt the increase in atmospheric CO2 concentrations. They can only complement efforts to reduce carbon emissions from the burning of fossil fuels. The Kyoto Protocol may have a profound influence on the forest sector, but its precise impacts will depend on which forest activities are included as eligible measures for climate change mitigation and what rules and standards are applied to potential projects. Opinions about the role of forestry within the Protocol's Clean Development Mechanism (CDM) are divided.

FIGURE 23

Estimates of carbon sources and the sink potential of various land use options

Note: For afforestation/reforestation, the potential rate assumes that 30 percent of suitable land is used; the maximum rate assumes that all available land is used. For reducing deforestation, the potential rate is based on the estimates of Brown et al., 1996; the maximum rate assumes a steady decline in tropical deforestation, with a complete halt after 50 years.

Opponents of forestry's inclusion in the CDM argue that incentives for carbon sequestration are likely to lead to uncontrolled investment in industrial-scale forest activities, with negative social and biological diversity consequences. Some observers fear that the availability of forestry as a low-cost means of achieving emissions reduction targets will divert investment from efforts to reduce emissions at their source. There are also concerns about the sustainability and measurability of forestry project impacts.

Proponents, however, see potential social, economic and biological diversity benefits arising from investment in high-quality conservation, agroforestry and sustainable forest management initiatives.

They argue that the additional economic (or carbon) value given to forests may provide a useful impetus to sustainable forest management efforts.

---

14 As of June 2001.

15 IPCC was set up in 1988 by the World Meteorological Organization and the United Nations Environment Programme. It provides the world community, but in particular the Parties to the FCCC, with scientific, technical and socio-economic information and advice related to human-induced climate change.

16 1 Gt is equivalent to 1 billion tonnes.

17 Data for carbon emissions resulting from land use change in the 1990s are not yet available.

18 This section is based on WCMC (in press). The information is reproduced here with the permission of the World Conservation Monitoring Centre.

19 Such changes could include the sudden release of methane from ocean deposits or the oxidation of northern forest soil carbon reserves, either of which would lead to accelerated warming or the slowing down of the North Atlantic thermohaline circulation, which could possibly lead to climate cooling.

20 AIJ projects are pilot projects carried out under the FCCC for testing and eval‎uating the feasibility of achieving the Convention's objectives through cooperative efforts between Parties to avoid, sequester or reduce greenhouse gas emissions.

21 In this context, a carbon offset is the amount of carbon withdrawn from the atmosphere by storage in vegetation and soil over an agreed period (the convention used by IPCC to calculate warming potential is 100 years) to compensate for the radiative forcing of an emission of a specified quantity of CO2 or another greenhouse gas.