IPCC Fourth Assessment Report: Climate Change 2007
Climate Change 2007: Working Group III: Mitigation of Climate Change Regional bottom-up assessments

Regional assessments comprise a variety of model results. On the one hand, these assessments are able to take into account the detailed regional specific constraints (in terms of ecological constraints, but also in terms of land owner behaviour and institutional frame).On the other hand, they also vary in assumptions, type of potential addressed, options taken into account, econometrics applied (if any), and the adoption of baselines. Thus, these assessments may have strengths, but when comparing and summing up, they have weaknesses as well. Some of these assessments, by taking into account institutional barriers, are close to a market potential.


The available studies about mitigation options differ widely in basic assumptions regarding carbon accounting, costs, land areas, baselines, and other major parameters. The type of mitigation options considered and the time frame of the study affect the total mitigation potential estimated for the tropics. A thorough comparative analysis is, therefore, very difficult. More detailed estimates of economic or market potential for mitigation options by region or country are needed to enable policy makers to make realistic estimates of mitigation potential under various policy, carbon price, and mitigation program eligibility rule scenarios. Examples to build on include Benitez-Ponce et al. (2007) and Waterloo et al. (2003), highlighting the large potential by avoiding deforestation and enhancing afforestation and reforestation, including bioenergy.

Reducing deforestation

Assumptions of future deforestation rates are key factors in estimates of GHG emissions from forest lands and of mitigation benefits, and vary significantly across studies. In all the studies, however, future deforestation is estimated to remain high in the tropics in the short and medium term. Sathaye et al. (2007) estimate that deforestation rates continue in all regions, particularly at high rates in Africa and South America, for a total of just under 600 million ha lost cumulatively by 2050. Using a spatial-explicit model coupled with demographic and economic databases, Soares-Filo et al. (2006) predict that, under a business-as-usual scenario, by 2050, projected deforestation trends will eliminate 40% of the current 540 million ha of Amazon forests, releasing approximately 117,000 ± 30,000 MtCO2 of carbon to the atmosphere (Box 9.1).

Reducing deforestation is, thus, a high-priority mitigation option within tropical regions. In addition to the significant carbon gains, substantive environmental and other benefits could be obtained from this option. Successfully implementing mitigation activities to counteract the accelerated loss of tropical forests requires understanding the causes for deforestation, which are multiple and locally based; few generalizations are possible (Chomitz et al., 2006).

Recent studies have been conducted at the national, regional, and global scale to estimate the mitigation potential (areas, carbon benefits and costs) of reducing tropical deforestation. In a short-term context (2008-2012), Jung (2005) estimates that 93% of the total mitigation potential in the tropics corresponds to avoided deforestation. For the Amazon basin, Soares- Filo et al. (2006) estimate that by 2050 the cumulative avoided deforestation potential for this region reaches 62,000 MtCO2 under a “governance” scenario (see Box 9.1).

Box 9.1 Deforestation scenarios for the Amazon Basin

An empirically based, policy-sensitive simulation model of deforestation for the Pan-Amazon basin has been developed (Soares-Filho et al., 2006) (Figure 9.7). Model output for the worst-case scenario (business-as-usual) shows that, by 2050, projected deforestation trends will eliminate 40% of the current 5.4 million km2 of Amazon forests, releasing approximately 117,000 MtCO2 cumulatively by 2050. Conversely, under the best-case governance scenario, 4.5 million km2 of forest would remain in 2050, which is 83% of the current extent or only 17% deforested, reducing cumulative carbon emissions by 2050 to only 55,000 MtCO2. Current experiments in forest conservation on private properties, markets for ecosystem services, and agro-ecological zoning must be refined and implemented to achieve comprehensive conservation. Part of the financial resources needed for these conservation initiatives could come in the form of carbon credits resulting from the avoidance of 62,000 MtCO2 emissions over 50 years.

Figure 9.7

Figure 9.7: Current carbon stocks for the Pan-Amazon and Brazilian Amazon (left bar) and estimates of cumulative future emission by 2050 from deforestation under BAU (business-as-usual) and governance scenarios.

Note: The difference between the two scenarios represents an amount equivalent to eight times the carbon emission reduction to be achieved during the first commitment period of the Kyoto Protocol.

Looking at the long-term, (Sohngen and Sedjo, 2006) estimate that for 27.2 US$/tCO2, deforestation could potentially be virtually eliminated. Over 50 years, this could mean a net cumulative gain of 278,000 MtCO2 relative to the baseline and 422 million additional hectares in forests. For lower prices of 1.36 US$/tCO2, only about 18,000 MtCO2 additional could be sequestered over 50 years. The largest gains in carbon would occur in Southeast Asia, which gains nearly 109,000 MtCO2 for 27.2 US$/tCO2, followed by South America, Africa, and Central America, which would gain 80,000, 70,000, and 22,000 MtCO2 for 27.2 US$/tCO2, respectively (Figure 9.5).

Figure 9.5

Figure 9.5: Cumulative carbon gained through avoided deforestation by 2055 over the reference case, by tropical regions under various carbon price scenarios

Source: Sohngen and Sedjo, 2006.

In a study of eight tropical countries covering half of the total forested area, Grieg-Gran (2004) present a best estimate of total costs of avoided deforestation in the form of the net present value of returns from land uses that are prevented, at 5 billion US$ per year. These figures represent costs of 483 US$ to 1050 US$/ha.

Afforestation and reforestation

The assumed land availability for afforestation options depends on the price of carbon and how that competes with existing or other land-use financial returns, barriers to changing land uses, land tenure patterns and legal status, commodity price support, and other social and policy factors.

Cost estimates for carbon sequestration projects for different regions compiled by Cacho et al., (2003) and by Richards and Stokes (2004) show a wide range. The cost is in the range of 0.5 US$ to 7 US$/tCO2 for forestry projects in developing countries, compared to 1.4 US$ to 22 US$/tCO2 for forestry projects in industrialized countries. In the short-term (2008-2012), an estimate of economic potential area available for afforestation/ reforestation under the Clean Development Mechanism (CDM) is estimated to be 5.3 million ha in Africa, Asia and Latin America together, with Asia accounting for 4.4 million ha (Waterloo et al., 2003).

Summing the measures, the cumulative carbon mitigation benefits (Figure 9.6) by 2050 for a scenario of 2.7 US$/tCO2 + 5% annual carbon price increment for one model are estimated to be 91,400 MtCO2; 59% of it coming from avoided deforestation. These estimates increase for a higher price scenario of 5.4 US$/tCO2 + 3%/yr annual carbon price into 104,800 MtCO2), where 69% of total mitigation comes from avoiding deforestation (Sathaye et al., 2007). During the period 2000-2050, avoided deforestation in South America and Asia dominate by accounting for 49% and 21%, respectively, of the total mitigation potential. When afforestation is considered, Asia dominates. The mitigation potential of the continents Asia, Africa and Latin America dominates the global total mitigation potential for the period up to 2050 and 2100, respectively (Figure 9.6).

Figure 9.6

Figure 9.6: Cumulative mitigation potential (2000-2050 and 2000-2100) according to mitigation options under the 2.7 US$/tCO2 +5%/yr annual carbon price increment

Source: Sathaye et al., 2007.

In conclusion, the studies report a large variety for mitigation potential in the tropics. All studies indicate that this part of the world has the largest mitigation potential in the forestry sector. For the tropics, the mitigation estimates for lower price ranges (<20 US$/tCO2) are around 1100 MtCO2/yr in 2040, about half of this potential is located in Central and South America (Sathaye et al., 2007; Soares Filho et al., 2006; Sohngen and Sedjo, 2006). For each of the regions Africa and Southeast Asia, this mitigation potential is estimated at 300 MtCO2/yr in 2040. In the high range of price scenarios (< 100 US$/tCO2), the mitigation estimates are in the range of 3000 to 4000MtCO2/yr in 2040. In the summary overviews in Section 9.4.4, an average estimate of 3500 is used, with the same division over regions: 875, 1750 and 875 for Africa, Latin and South America, and Southeast Asia, respectively. The global economic potential for the tropics ranges from 1100 to 3500 MtCO2/yr in 2040 (Table 9.6).

OECD North America

Figure 9.8 shows the technical potential of management actions aimed at modifying the net carbon balance in Canadian forests (Chen et al., 2000). Of the four scenarios examined, the potential was largest in the scenario aimed at reducing regeneration delays by reforesting after natural disturbances. The second largest estimate was obtained with annual, large-scale (125 million ha) low-intensity (5 kg N/ha/yr) nitrogen fertilization programmes. Neither of these scenarios is realistic,

Figure 9.8

Figure 9.8: OECD North America: technical potential for the forest sector alone for Canada (left, sink is positive) and the economic potential (at 15 US$/tCO2eq in constant real prices) in the agriculture and forestry sector in the USA (right) Left: Chen et al., 2000; right: US-EPA, 2005

however, but can be seen as indications of the type of measures and impact on carbon balance (as described by Chen et al., 2000). Chen’s measures sum up to a technical potential of 570 MtCO2/yr. Based on the assumption that the economic potential is about 10% of technical potential (see Section for carbon prices 20 US$/tCO2), the economic potential can be “guesstimated” at around 50-70 MtCO2/yr (Table 9.6).

Other studies have explored the potential of large-scale afforestation in Canada. Mc Kenney et al. (2004) project that at a carbon price of 25 US$/tCO2, 7.5 million ha of agricultural land would become economically attractive for poplar plantations. Economic constraints are contributing to the declining trend in afforestation rates in Canada from about 10,000 ha/yr in 1990 to 4,000 ha/yr in 2002 (White and Kurz, 2005).

For the USA, Richards and Stokes (2004) reviewed eight national estimates of forest mitigation and found that carbon prices ranging from 1 to 41 US$/tCO2 generated an economic mitigation potential of 47-2,340 MtCO2/yr from afforestation, 404 MtCO2 from forest management, and 551-2,753 MtCO2/yr from total forest carbon. Sohngen and Mendelsohn (2003) found that a carbon programme with prices rising from 2 US$/tCO2 to 51 US$/tCO2 during this century could induce sequestration of 122 to 306 MtCO2/yr total carbon sequestration, annualized over a 100-year time frame.

US EPA (2005) present that, at 15 US$/tCO2, the mitigation potential of afforestation and forest management (annualized) would amount to 356 MtCO2/yr over a 100-year time frame. At 30 US$/tCO2, this analysis would generate 749 MtCO2 annualized over 100 years. At higher prices and in the long term, the potential was mainly determined by biofuels. With the mitigation potential given above for Canada, the OECD North America sums to a range of 400 to 820 MtCO2/yr in 2040 (Table 9.6).


Most assessments shown (Figure 9.9) are of the carbon balance of the forest sector of Europe’s managed forest as a whole[4]. Additional effects of measures were studied by Cannell (2003), Benitez-Ponce et al. (2007), EEA (2005), and Eggers et al. (2007). Karjalainen et al, (2003) present a projection of the full sector carbon balance (Figure 9.9). Eggers et al. (2007) presents the European forest sector carbon sink under two global SRES scenarios, and a maximum difference between scenarios of 197 MtCO2/yr in 2040. Therefore, an additionally achievable sink of 90 to 180 MtCO2/yr was estimated (Table 9.6). Economic analyses were not only done; country studies were done, for example, Hoen and Solberg (1994) for Norway. New European scale economic analyses may be available from the INSEA[5] project, MEACAP project[6], and Carbo Europe [7].

Issues in European forestry where mitigation options can be found include: afforestation of abandoned agricultural lands; bioenergy from complementary fellings; and forest management practices to address carbon saturation in older forests. Furthermore, management of small now under-managed woodlands represent a potential (Viner et al., 2006) and also in combination with adaptation measures in connecting the fragmented nature reserves (Schröter et al., 2005).

Figure 9.9

Figure 9.9: European forest sector carbon sink projections for which various assumptions on implementation rate of measures were made

Note: positive = sink.

Russian Federation

The forests of the Russian Federation include large areas of primary (mostly boreal) forests. Most estimates indicate that the Russian forests are neither a large sink nor a large source. Natural disturbances (fire) play a major role in the carbon balance with emissions up to 1,600 MtCO2/yr (Zhang et al., 2003). Large uncertainty surrounds the estimates for the current carbon balance ((Shvidenko and Nilsson et al., 2003). For the decade 1990-2000, the range of carbon sink values for Russia is 350-750 MtCO2/yr (Nilsson et al., 2003; Izrael et al., 2002). A recent analysis estimated the net sink in Russia at 146-439 MtCO2/yr at present (Sohngen et al., 2005). They projected this baseline to be about 257 MtCO2 per year in 2010, declining to a net source by 2030 as younger forests mature and are harvested. They estimated the economic potential in Russia of afforestation and reforestation at 73-124 MtCO2/yr on average over an 80-year period, for a carbon price of 1.9-3.55 US$/tCO2, and 308-476 MtCO2/yr at prices of more than 27 US$/tCO2 (Figure 9.10). Based on these estimates, the estimated economic mitigation potential would be between 150 and 300 MtCO2/yr in the year 2040 (Table 9.6).

Figure 9.10

Figure 9.10: Russian Federation forest sector carbon sink projections, with assumptions regarding implementation rates differing in the various studies

Note: positive = sink.

OECD Pacific

Richards and Brack (2004) used estimates of establishment rates for hardwood (short and long rotation) and softwood plantations to model a carbon account for Australia’s post-1990 plantation estate. The annual sequestration rate in forests and wood products together is estimated to reach 20 MtCO2/yr in 2020.

New Zealand reached a peak in new planting of around 98,000 ha in 1994 and estimates of stock changes largely depend on afforestation rates (MfE, 2002). If a new planting was maintained at 40,000 ha/yr, the stock increase in forests established since 1990 (117 MtCO2 cumulative since 1990) is estimated to offset all increases in emissions in New Zealand since 1990. The total stock increase in all forests would offset all emissions increases until 2020.

However, the current new planting rate has declined to 6,000 ha and conversion of 7,000 ha of plantations to pasture has led to net deforestation in the year to March 2005 (MAF, 2006). As a result, the total removal units anticipated to be available during the first commitment period dropped to 56 MtCO2 in 2005 (MfE, 2005). Trotter et al. (2005) estimate New Zealand has approximately 1.45 million ha of marginal pastoral land suitable for afforestation. If all of this area was established, total sequestration could range from 10 to 42 MtCO2/yr. This would lead to a removal of approximately 44 to 170 MtCO2 cumulative by 2010 at 13 US$/tCO2.

In Japan, 67% of the land is covered with forests including semi-natural broad-leaved forests and planted coniferous forests mostly. The sequestration potential is estimated in the range of 35 to 70 MtCO2/yr (Matsumoto et al., 2002; Fang et al., 2005), and planted forests account for more than 60% of the carbon sequestration. These assessments show that there is little potential for afforestation and reforestation, while forest management and practices for planted forests including thinning and regeneration are necessary to maintain carbon sequestration and to curb saturation. In addition, there seems to be large potential for bioenergy as a mitigation option.

These three countries for the region lead to an estimate of potential in the range of 85 to 255 MtCO2/yr in 2040 (Table 9.6).

Non-annex I East Asia

East Asia to a large extent formed by China, Korea, and Mongolia has a range of forest covers from a relatively small area of moist tropical forest to large extents of temperate forest and steppe-like shrubland. Country assessments for the forest sector all project a sink ranging from 75 to 400 MtCO2/yr (Zhang and Xu, 2003). Given the large areas and the fast economic development (and thus demand for wood products resulting in increased planting), the additional potential in the region would be in the high range of the country assessments at 150 to 400 MtCO2/yr (Table 9.6). Issues in forestry with which the carbon sequestration goal can be combined sustainably are: reducing degradation of tropical and dry woodlands; halting desertification of the steppes (see Chapter 8); afforestation; and bioenergy from complementary fellings.

  1. ^  Europe here excludes the European part of Russia.
  2. ^  www.iiasa.ac.at/Research/FOR/INSEA/index.html?sb=19
  3. ^  www.ieep.eu/projectMiniSites/meacap/index.php
  4. ^  www.carboeurope.org/