8 Carbon budgets of tropical forest ecosystems in Southeast Asia: implications for climate change

Rodel D. Lasco^[12] and Florencia B. Pulhin^[13]

ABSTRACT

Terrestrial ecosystems have an important role to play in the global carbon (C) cycle. Tropical forests in Southeast Asia are constantly changing as a result of harvesting and conversion to other land covers. In the last few years, research on the C budgets of forest ecosystems as a result of these changes has intensified in the region. Natural forests in SE Asia typically contain a high C density (up to 500 Mg ha^-1). Logging activities result in at least a 50 percent decline in C density of the forest. On the other hand, complete deforestation (conversion to grassland and annual crops) leads to C densities of generally less than 40 Mg ha^-1. Conversion to tree plantations and other woody perennial crops also reduces C density to less than 50 percent of the original C in the forests. Finally, the implications of the above changes in C budgets to climate change and the CDM are discussed.

INTRODUCTION

There is considerable interest on the role of terrestrial ecosystems in the global carbon (C) cycle. It is estimated that about 60 Pg^[14] C is exchanged between terrestrial ecosystems and the atmosphere every year, with a net terrestrial uptake of 0.7 ± 1.0 Pg C (Schimell et al. 1995). The world’s tropical forests, which cover 17.6 M km², contain 428 Pg C in vegetation and soils. On the other hand, land use, land-use change and forestry (LULUCF) activities, mainly tropical deforestation, are significant net sources of CO₂, accounting for 1.6 Pg y^-1 out of the total anthropogenic emissions of 6.3 Pg y^-1 (Houghton 1996, Watson et al. 2000). However, tropical forests have the largest potential to mitigate climate change amongst the world’s forests through conservation of existing C pools (e.g. reduced impact logging), expansion of C sinks (e.g. reforestation, agroforestry), and substitution of wood products for fossil fuels (Brown et al. 1996). In tropical Asia, it is also estimated that forestation, agroforestry, regeneration and avoided deforestation activities have the potential to sequester 7.50, 2.03, 3.8-7.7, and 3.3-5.8 Pg C respectively between 1995 and 2050 (Brown et al. 1996).

In spite of their importance to the C cycle, there is little information on the effects of land-use change and management activities on the C budgets of forest ecosystems in the tropics. For example, one of the major research needs identified in the Second Assessment Report of the IPCC is how different silvicultural and other management practices would affect the C dynamics in forests (Brown et al. 1996). In the last few years, research on C stocks and dynamics in forest ecosystems in Southeast Asia has intensified and new data have been generated (Lasco and Pulhin 2000, Murdiyarso 2000, Lasco et al. 2001, Lasco 2002). This paper attempts to review the available information on C budgets of forest ecosystems in Southeast Asia in response to land-use change and management activities such as harvesting and deforestation.

FOREST LAND-USE CHANGE IN SOUTHEAST ASIA AND THE C CYCLE

Rate of deforestation and land-use/cover change in Southeast Asia

The last few decades have seen massive deforestation and land-use/cover change in the tropics and Southeast Asia was no exception. In fact, tropical deforestation is the dominant change in land use in the tropics (Lugo and Brown 1992). Deforestation rates in tropical Asia were estimated to be 2.0 M ha in 1980 and 3.9 M ha in 1981-1990 (Brown 1993). In Southeast Asia, the 1990 annual deforestation rate was about 2.6 M ha y^-1 (Table 1).

Table 1. 1990 annual deforestation estimates for countries in Southeast Asia (from Trexler and Haugen 1994)

There exist varying estimates of the rates of deforestation for each country, partly because of different time frames and sources of data. This is illustrated in the case of the Philippines where deforestation rates have fluctuated in the last 100 years with an average of about 150 000 ha y^-1 (Table 2). The forest area in Thailand declined from 28.03 M ha to 13.35 M ha between 1961 and 1993 (Boonpragob 1998), an average loss of 445 000 ha y^-1, much higher than the estimate in Table 1. However, a recent GHG study in the country, estimated deforestation rate at 150 000 ha y^-1 in 1992-93 (ALGAS 1998 as cited by Macandog 2000). The initial National Communication of Indonesia estimated the rate of forest conversion at about 800 000 y^-1 (State Ministry for Environment 1999); other estimates place it at 1 M ha in the early 1990s and 721 000 ha in 1994 (Macandog 2000).

Table 2. Deforestation rates in the Philippines in the 20th century

Forest land-cover change dynamics in Southeast Asia^[15]

In the last century, commercial logging in the Philippines has been the main cause of conversion of old-growth (primary) forests to secondary forests. In addition, small-scale swidden farming is also deemed responsible for the formation of secondary forests (Kummer 1992). Since 1900, the Philippines has lost about 15 M ha of tropical forests. It can be presumed that these were first converted to secondary forests before being totally denuded.

Secondary forests could be converted to the following land uses: upland farms, pasture areas, brushlands and tree plantations. Conversion to upland farms is typically done by farmers who follow at the heels of loggers. Logged-over areas are easy to clear because the largest trees have been removed and logging roads provide easy access. Upland farms may revert back to secondary forests through fallow. Forest fallows are more often associated with indigenous peoples. Several indigenous fallow systems have been documented in the Philippines (Hanunos of Mindoro Tagbanua of Palawan, etc.). However, upland farms of migrant farmers hardly, if ever, revert back to forests as they are continuously cultivated until the soils are very degraded. Most grasslands in the Philippines are formed in this manner.

Clearing of secondary forest areas for pasture could also have happened in the past; however, it is more likely that pasture areas were former upland farms. When abandoned, pasture areas remain grasslands because of very poor soils and regular burning. If fire is controlled, studies and observations have shown that grasslands can return to secondary forests through natural succession (Friday et al. 1999).

Secondary forests could become brushlands as a result of continuous cutting of trees, mostly illegal. Brushlands contain about 20 percent forest cover or less. If further denuded, they could become grassland areas. However, if disturbance ceases, they revert back to secondary forests. Finally, secondary forests could also be converted to tree plantations. This is not allowed anymore at present, but could have been significant in the past. Tree plantations rarely return back to natural forests. The case of the Philippines is not unique. In many countries in the region, forest conversion starts with selective logging and ends with degraded pastures and grasslands (Detwiler and Hall 1988).

C budgets of forest ecosystems and their potential for C sequestration

Tropical forests contain a significant amount of C in the biomass, necromass and in the soil. In tropical Asia, 41-54 Pg C and 43 Pg C are found in vegetation and soils respectively (Dixon et al. 1994). Annual C flux from tropical Asian forests is estimated at -0.50 to -0.90 Pg y^-1.

In terms of their potential to mitigate carbon from the atmosphere, the amount of C that could be sequestered in above-ground biomass of the technically suitable present forest lands in Asia is estimated to be about 15 Pg or an average of 88 Mg C ha^-1 (Iverson et al. 1993). Another study showed that Southeast Asian countries have the potential to sequester 9.0 Pg to 21.0 Pg of C from 1995 to 2050 in 66 M ha through forest regeneration, farm forestry and plantation development (Table 3).

Table 3. Potential C sequestration of regeneration, farm forestry and plantation development activities in Southeast Asia (from Trexler and Haugen 1994)

CARBON BUDGETS FOLLOWING LOGGING OPERATIONS IN NATURAL FORESTS

Carbon stocks and rate of sequestration of natural forests in Southeast Asia

Using GIS, Brown et al. (1993) estimated that in 1980 the average C density for tropical forests in Asia was 144 Mg ha^-1 of actual biomass, and 148 Mg ha^-1 in soils (up to 100 cm) which corresponds to total estimates of 42 and 43 Pg respectively, for the whole continent. It was noted that C densities and pools in vegetation and soil varied widely by ecofloristic zone and country. Actual biomass C densities range from less than 50 to more than 360 Mg C ha^-1 with most forests having 100-200 Mg C ha^-1. The higher biomass is in Borneo and Irian Jaya (Indonesia) while there was lower biomass C in forests in India and Thailand. C densities in soils range from 60 to 160 Mg ha^-1.

A similar study by Iverson et al. (1993) reported an average maximum C stock in present forest lands in tropical Asia of 185 Mg C ha^-1 with a range of 25 to more than 300 Mg C ha^-1. On the other hand, Palm et al. (1986), as reported by Houghton (1991), found out that forests in tropical Asia have C density of 40-250 Mg ha^-1 and 50-120 Mg ha^-1 in vegetation and soils respectively (Table 4). Brown et al. (1991) reported that Southeast Asian forests have a biomass range of 50-430 Mg ha^-1 (25-215 Mg C ha^-1) and >350-400 Mg ha^-1 (175-200 Mg C ha^-1) before human incursion. For national GHG inventories, the IPCC (1996) recommends a default value of 275 Mg C ha^-1 for wet forests in Asia.

Table 4. Carbon in vegetation and soils of forest ecosystems in tropical Asia (from Palm et al. (1986) as cited by Houghton (1991))

There are limited data on carbon densities of natural forests in specific Southeast Asian countries. Most of the recent studies have been reported for Indonesia and the Philippines. These studies are largely based on the use of allometric equations to estimate tree biomass (e.g. equations from Brown (1997)). Indonesian forests have C density ranging from 65 to 390 Mg C ha^-1 (Table 5).

Table 5. Biomass and C density of natural forests in Indonesia

On the other hand, recent studies report that Philippine natural forests contain 86-201 Mg C ha^-1 (Table 6). The IPCC Revised Guidelines (IPCC 1996) estimates that old-growth forests in the Philippines contain 370-520 Mg ha^-1 of above-ground biomass equivalent to about 185-260 Mg C ha^-1 at 50 percent C content.

Table 6. Biomass and C density of natural forests in the Philippines

For Thailand, it is reported that the various types have a C density ranging from 72 to 182 Mg C ha^-1 (Table 7). A similar data set for Thailand is presented in Table 8. These C data are used in the national GHG inventory reports of the country.

Table 7. Above-ground biomass and C density of various forest types in Thailand (Boompragob 1998)

(from various sources; C density calculated based on 50 percent C content)

EGF - tropical evergreen forest; MDF - mixed deciduous forest; DDF - dry dipterocarp forest; PF - pine forest; MF - mangrove forest

Table 8. Biomass and C density of forests in Thailand (from Macandog 2000b)

C density calculated based on 50 percent C density.

Malaysian forests have C density ranging from 100 to 160 Mg C ha^-1 and from 90 to 780 Mg C ha^-1 in vegetation and soils respectively (Table 9). Cairns et al. (1997), citing various sources, reported that mature lowland forests have above-ground biomass and root density of 431 and 43 Mg ha^-1 respectively, equivalent to 216 and 22 Mg C ha^-1 respectively.

Table 9. Carbon density of various forest types in Malaysia (Abu Bakar 2000)

Carbon budgets of forest ecosystems after logging operations

Natural forests in the Southeast Asian region have been one of the world’s foremost sources of tropical hardwoods. Logging activities are therefore dominant in many countries. As discussed earlier, logging operation is primarily responsible for the conversion of primary forests to secondary forests. Destructive logging and subsequent agricultural conversion have vastly depleted natural forests and left millions of hectares of degraded lands in each country. Some countries, notably Thailand and the Philippines, have banned logging operations in primary forests.

In general, logging leads to a reduction of C stocks in the forest as biomass is reduced by the extraction of wood. C is released upon the decomposition or burning of slash and litter. However, regenerating trees sequester C back to biomass over time. In general, the biomass and C density of tropical forests in Asia decline by 22-67 percent after logging (Table 10).

Table 10. Biomass and C (in parenthesis) density (Mg C ha^-1) of tropical forests in Asia (Brown and Lugo 1984)

In the Philippines, we studied the carbon density of logged-over forest plots with varying ages after logging (Lasco et al. 2000). Right after logging, C density declined to about 50 percent of the mature forest (198 Mg C ha^-1). There was no other similar study in other Southeast Asian countries which tracks the decline of C density after logging. However, measurements have been taken in logged-over forests which could be compared to primary forests in those countries. In Indonesia, estimates of C density of logged-over forests range from 38 to 75 percent of the original forest cover (Table 11).

Table 11. Carbon density after logging of Indonesian forests

As can be gleaned from above, logging is typically a very destructive practice. In Malaysia, extracting 8-15 trees (80 m³; ca. 22 Mg C ha^-1) damaged as many as 50 percent of the remaining trees (Putz and Pinard 1993). Out of the initial 348 Mg C ha^-1, 95 Mg C ha^-1 are transformed to necromass, which eventually releases its C via decomposition. In the Philippines, for every tree cut greater than 75 cm dbh, 1.5 and 2.6 trees are damaged in favourable and unfavourable conditions respectively (Weidelt and Banaag 1982).

However, numerous studies have shown that logging damage can be significantly reduced by directional felling and well-planned skid trails (Putz and Pinard 1993). These practices are collectively known as reduced-impact logging (RIL). The effect of RIL on C conservation has been thoroughly investigated in a study conducted in Sabah, Malaysia, as reported by Pinard and Putz (1997, 1996). The rest of this section is based on their reports.

The biomass and C density are very similar before logging operations (Table 12). About one year after logging, forest areas logged conventionally and under RIL contained 44 percent and 67 percent of their pre-logging biomass respectively (Table 13). The C density of RIL was 88 Mg C ha^-1 higher than conventional logging. In terms of logging damage, in RIL about 27 percent of trees >10 cm dbh were damaged and about 19 percent were dead within the first year after logging. In contrast, in conventional logging, about 54 percent were damaged and about 46 percent died. Expectedly, there was 86 Mg C ha^-1 less necromass in RIL compared to conventional logging (Table 14), which will translate to lower CO₂ emissions from decomposition.

After logging, biomass C is projected to decline in both areas for 2-6 years because of high mortality rates and decay of logging debris. Following stabilization of mortality rates, biomass accumulation will be greater in RIL areas. Indeed, they will become net sinks in fewer years than conventional logging. Modelling showed that during the 40-year project life span, about 90 Mg C ha^-1 will exist in forest biomass due to RIL. Of these, 55 percent will be present 10 years after logging due to less damage.

Table 12. Above-and below-ground biomass of dipterocarp forest in Ulu Segama Forest Reserve, Sabah, Malaysia, before logging (SD, number of plots or logging units)

C content= 49.2%.

Table 13. Above-and below-ground biomass (and necromass) for the two logging treatment areas 8 -12 months after logging; for coarse roots, three months after logging. Means (Mg ha^-1) presented with SD and N noted parenthetically. For trees, vines, and butt root mass, SD describes variation among logging units and does not incorporate errors in biomass equations

Table 14. Mean (SD) Mg biomass per ha converted to necromass. SD describes variation among four logging units and does not incorporate error in biomass equations

CARBON BUDGETS FOLLOWING CONVERSION FROM FOREST TO NON-FOREST COVER

Impact of deforestation on carbon budgets

As discussed earlier, deforestation is a major land-use change in Southeast Asia. There are no studies that directly track the change in C budget through the deforestation process. However, there are studies that have quantified the C stocks in deforested lands, typically covered with grasslands or annual crops.

In Indonesia, various reports show that above-ground C density in grasslands and shifting cultivation areas is less than 40 Mg C ha^-1 (Table 15). In the Philippines, grassland and crop lands contain 3.1 to 13.1 Mg C ha^-1. In both countries, these are vastly lower than the C density in the natural forests they replaced.

Soil organic carbon (SOC) may also be affected by the change in land use. However, no data are available for Southeast Asia. In general, many studies have shown that continuously cultivated systems have lower SOC than adjacent forests (Lugo and Brown 1993). However, pasture areas can accumulate as much C in the soil as adjacent natural forests.

Conversion to tree plantations and perennial crops

Natural forest areas, usually after commercial logging, can be converted to plantations of forest trees or perennial crops. This land-use change is expected to reduce C stocks. There are no studies that directly measure the change of C stocks as a result of this change through time. However, by comparing the C stocks of the resulting land use with the C stocks of a natural forest, we can have an idea of the magnitude of change. This kind of comparison is of course preliminary as the C stocks vary with age of the plantation and the site characteristics.

In a multi-country study, tree and agricultural plantations have C stocks that are 7-51 percent lower than natural forests (Table 16). Similarly, another study in Indonesia showed that agroforestry and plantation farms had C stocks that are 4-66 percent lower than that of an undisturbed forest (Table 17). These data also show how C stocks vary with the age of rubber plantation, with older rubber agroforests having almost seven times more C than a 5-year-old plantation.

Table 15. Above-ground biomass density of grasslands and annual crops in Indonesia and the Philippines

Table 16. C density of tree and agricultural plantations in the Philippines and Indonesia

Table 17. C stocks of agroforestry and plantation farms in Jambi and Lampung, Indonesia (C density data from Sitompul and Hairiah 2000)

In a lowland peneplein in Indonesia, rubber and oil palm plantations were estimated to contain 36-46 percent of the C of the natural forest (Table 18) while various land-cover types in Indonesia were estimated to contain 14-63 percent of the C density of a natural forest (Table 19). These land-cover types are briefly described below:

• rubber jungle: rubber and secondary vegetation; Jambi Province (Prasetyo et al. 2000);

• home gardens: cultivation of annual and perennial crops on the same piece of land near the house (Sitompul and Hairiah 2000). The data presented is from Malang;

• oil palm: 19-year-old plantation; soil organic C 48.3 Mg ha^-1 (Tjitrosemito and Mawardi 2000);

• cinnamon: part of 10-ha demo plots in Sarolangun Bangko district; in vast areas of degraded grassland (recovery); 7-year-old plantation (Gintings 2000);

• Acacia mangium: 9-year-old plantation (Siregar et al. 1998).

Table 18. Time-averaged C stocks for lowland peneplein in Indonesia (above-ground biomass and top 30 cm soil) (age and C data from Noordwijk et al. 2000)

Table 19. C density of various land cover in Indonesia

* Natural forest assumed to contain 254 Mg C ha^-1 based on Noordwijk et al. 2000.
** Calculated by Sitompul & Hairiah (2000).

In Mindanao, Philippines, tree plantations of fast-growing species contain 3-45 percent of the C of a natural dipterocarp forest (Table 20). On the other hand, a mature coconut plantation in Leyte Province contains 86 Mg C ha^-1 in above-ground biomass (Lasco et al. 1999), which is about 43 percent of a natural forest in the same area (259 Mg C ha^-1; Lasco et al. 2001).

Table 20. C density of tree plantations in Mindanao, Philippines

* Harvested 20 years ago.
Biomass data from Kawahara (1981); C content assumed to be 45 percent (Lasco and Pulhin 2000)

Agroforestry systems have been widely promoted as an alternative technology to slash-and-burn farming. They involve planting of tree and perennials in conjunction with agricultural crops. Various forms of agroforestry exist in the Philippines (Lasco and Lasco 1989). A Leucaena leucocephala fallow field in Cebu, Philippines, has a mean C density of 16 Mg C ha^-1 during its 6-year cycle (Table 21). This is very low compared to natural forests in the country. A coconut-based multistorey system in Mt. Makiling has a C density in AGB of 39 Mg C ha^-1 (Zamora 1999) which is only about 15 percent of the C of adjacent natural forests.

Table 21. C density and MAI of a Leucaena leucocephala fallow field in Cebu, Philippines (from Lasco and Suson 1999)

Means in a column with the same letter are not significantly different using DMRT at 0.05.

In conclusion, it appears that tree and perennial crop plantations typically have C stocks in above-ground biomass that are less than 50 percent of that of natural forests they replace. In the process of converting natural forests to agricultural and tree plantations, burning is often used for site preparation. In Indonesia, changes in C stocks during land clearing from old jungle rubber/secondary forest for replanting rubber varied depending on whether burning is used (Noorwijk et al. 2000). “Slash and burn” lost 66 percent C from ca. 80 to 25 Mg C ha^-1 while “slash and mulch” (no burning) lost only 20 percent C (ca. from 110 to 90 Mg ha^-1). In north Lampung, the biomass declined from 161 Mg ha^-1 to 46 Mg ha^-1 because of burning (Hairaih et al. 1999). This is equivalent to a loss of about 58 Mg C ha^-1.

Once tree and perennial crop plantations have been established, they begin to accumulate C. Noordwijk et al. (2000) reported C accumulation rate of 2.5 Mg C ha^-1y^-1 in natural fallows (secondary forests), agroforests and more intensive tree-crop production systems in Indonesia. An example of these is jungle rubber system (Hairiah and Sitompul 2000).

Table 22 shows the rate of annual C accumulation by various forest plantations as used in the first Indonesian national communication to the UN Framework Convention on Climate Change (UNFCCC). It ranges from 0.50 to 12.50 Mg ha^-1y^-1.

Table 22. Annual C accumulation rate of various forest plantations used in the national GHG inventory of Indonesia (State Ministry of Environment 1999)

A 7-y-old cinnamon plantation in Indonesia accumulates C at the rate of 4.49 to 7.10 kg C tree^-1 (Table 23). In the Philippines, commercial tree plantations of fast-growing species sequestered C at the rate of 0.50-7.82 Mg C ha^-1 y^-1(Table 24). The next section also presents estimates of C density and rate of sequestration of reforestation/afforestation species.

Table 23. Rate of biomass and C accumulation (in kg) of a 7-y-old cinnamon plantation in Indonesia (Gintings 2000)

Table 24. MAI of biomass and carbon of tree plantations in Mindanao, Philippines

* Harvested 20 years ago; assumed to be 100 years old.
Biomass data obtained by destructive sampling (Kawahara et al. 1981).

Soil organic matter/carbon (SOM/SOC) is also affected by the change in land use. C in the soil is a significant pool. It has the longest residence time among organic C pools in the forest (Lugo and Brown 1993). However, the exact effect of land-use change on SOC is largely unknown in tropical forests, specially the rates and direction of change. Below are the available data in the Southeast Asian region, mainly from Indonesia and some from the Philippines.

In north Lampung, Indonesia, the total LUDOX fraction of SOC was reduced by 70-80 percent under degrading situations (burnt Imperata; sugarcane with burning; forest plantation with bulldozer for land clearing) in the top soil (0-5 cm), 8-10 years after converting the forest without effort of maintaining SOM (Murdiyarso et al. 1996). The methods of forest conversion also has big impact on SOM: slash-and-burn practices on forest plantation reduced total LUDOX fraction about 50 percent which may be due to washing away of the SOM during high intensity rainfall. On the other hand, clearing the forest using bulldozer reduced LUDOX fraction by about 70 percent.

Planting fast-growing species like P. falcataria increased LUDOX fraction at 0-5 cm by about five times higher than forested area. However, a mixed plantation of P. falcataria and A. magium reduced total LUDOX fraction up to 50 percent compared to forest. In the Rantau Pandan site, a newly developed cinnamon plantation reduced SOM by 30 percent while cassava plots increased SOM. In the Sitiung site, Imperata grass with regular burning after years of intensive cultivation also had great reduction of SOM. In another study, tree plantations in Indonesia also have lower SOC density than natural forests (Table 25).

In the Philippines, a coconut plantation was found to have about half the SOC density of a natural forest (111 Mg C ha^-1 vs 191 Mg C ha^-1) (Lasco et al. 1999).

Table 25. SOC at various depth and land use (Siregar and Gintings 2000)

IMPLICATIONS FOR CLIMATE CHANGE AND THE CDM

Land use, land-use change and forestry (LULUCF) activities, mainly tropical deforestation, are significant net sources of CO₂, accounting for 1.6 Pg y^-1 out of the total anthropogenic emissions of 6.3 Pg y^-1 (Houghton et al. 1996, Watson et al. 2000). The preceding discussion shows how logging activities, deforestation and land-use change affect the C stocks of tropical forests in SE Asia. Lowering of C stocks in terrestrial ecosystems means a corresponding increase of C emissions to the atmosphere.

Significant C is emitted to the atmosphere as a result of forest disturbance and clearing. As expected, deforestation causes the highest C emissions, more than 90 percent of the above-ground C stocks of a natural forests being lost. If the land is not reforested, these losses become permanent addition to the CO₂ concentration in the atmosphere. Logging also results in a loss of about 50 percent of C stocks. However, the C could be reabsorbed if the forests are allowed to regenerate. The biomass may not reach the level of the primary forest if there is overcutting or premature cutting. Conversion of natural forests to plantations will also increase C emissions as the forest is cleared. The C could also be reabsorbed as plantation crops grow. However, the C stocks are usually lower in plantations compared to the natural forests they replace.

In 1997, during the Third Conference of Parties (COP), the Kyoto Protocol was drafted, which is the first international agreement that places legally binding limits on GHG emissions from developed countries (UNFCCC 1997). The Protocol also provides for flexible mechanisms to meet carbon reduction obligations. The most relevant to developing countries is the Clean Development Mechanism (CDM) contained in Article 12. Essentially, the CDM allows Annex 1 (developed) countries to meet their carbon reduction quota via activities in developing countries (non-Annex 1 countries). Two forestry activities are allowed under the first commitment period: reforestation and afforestation. The CDM provides a way for developing countries to be more actively involved in the mitigation of GHG in the atmosphere, short of actual reduction commitments. But perhaps more importantly in the short term, developing countries stand to benefit from the CDM through investment in flow and technology transfer that will support their respective sustainable development agenda (Frumhoff et al. 1998).

The CDM offers an opportunity for SE countries with wide areas of barren lands to generate resources for their reforestation and thereby reabsorbed the C emitted from these lands due to deforestation. In addition, many social and environmental co-benefits could accrue as a result of reforestation activities. For example, the Philippines, with its wide areas of land needing reforestations stands to benefit in the CDM, should it decide to participate. There are anywhere from 2 to 9 M ha of denuded and degraded upland areas that need immediate rehabilitation (Lasco and Pulhin 2000). These areas were former tropical forests but are now mainly grasslands, brushlands and cultivated farms. At the present rate or reforestation (less than 100 000 ha y^-1), it will take more than 100 years to fully rehabilitate these areas. In addition, up to 19 million people are living in the uplands, half of whom rely on some form of shifting cultivation. By reforesting these lands, the country could potentially reap many co-benefits: income-generation, soil conservation, watershed rehabilitation, biodiversity conservation, etc.

CONCLUSION

On the basis of the foregoing review of C budgets with harvesting and land-cover change, the following conclusions emerge:

• C density in above-ground biomass declines by at least 50 percent after logging.

• Deforested areas covered with grasses and annual crops have C density less than 40 Mg C ha^-1, much lower than natural forests.

• Conversion of natural forests to tree plantations and perennial crops reduces C density by at least 50 percent relative to natural forests.

• Most studies conducted to estimate C density relied on allometric equations derived globally rather from within the country.

• SOC declines with deforestation and conversion of natural forests to plantations.

• There are still limited data on the C dynamics of tropical forests in Southeast Asia. Most of the existing data are on above-ground biomass C; there are less data on below-ground C in roots and soil.

In spite of the rise of available information in the last few years, there is clearly much more that needs to be done. It has to be noted that many research results are found in “grey literature”. In addition, many of the biomass and C data are based on extrapolation using allometric equations (mainly from Brown (1997)) rather than primary data collection. The following research topics need to be further pursued:

• generation of country-specific allometric equations for biomass and C density;
• assessment of C dynamics associated with key land-use/cover change;
• comprehensive C stocks assessment of LULUCF activities including above-and below-ground biomass and soils;
• effects of silvicultural treatments and management practices on C budgets of forest ecosystems.

BIBLIOGRAPHY

Abu Bakar, R. 2000. Carbon economy of Malaysian jungle/forest and oil palm plantation. Paper presented at the Workshop on LUCC and Greenhouse Gas Emissions Biophysical Data. 16 December 2000. Institute Pertanian Bogor, Indonesia.

ALGAS. 1998. Asia Least-Cost Greenhouse Gas Abatement Strategy (ALGAS) - executive summary. Asian Development Bank. Global Environment Facility, United Nations Development Programme. Manila, Philippines.

Boonpragob, K. 1998. Estimating greenhouse gas emission and sequestration from land use change and forestry in Thailand. In T.B. Moya, ed. Greenhouse gas emissions, aerosols, land use and cover changes in Southeast Asia. Pp. 18-25. Southeast Asia Regional Committee, Bangkok, Thailand.

Brown, S. 1993. Tropical forests and the global carbon cycle: the need for sustainable land-use systems. Agriculture, Ecosystems and Environment 46: 31-44.

Brown, S. 1997. Estimating biomass and biomass change of tropical forest: a primer. Forestry paper 134. Rome, FAO.

Brown, S., Gillespie, A.J.R. & Lugo, A.E. 1991. Biomass of tropical forests of South and Southeast Asia. Can. For. Res. 21: 111-117.

Brown, S., Iverson, L.I., Prassad, A. & Liu, D. 1993. Geographical distribution of carbon in biomass and soils of tropical Asian forests. Geocarto International 4: 45-59.

Brown, S. & Lugo, A.E. 1984. Biomass of tropical forests: a new estimate based on forest volumes. Science 223: 1290-1293.

Brown, S., Masera, O. & Sathaye, J. 2000. Project-based activities. In R.T. Watson, I.R. Noble, B. Bolin, N.H. Ravindranath, D.J. Verado & D.J. Dokken, eds. Land use, land-use change and forestry, Chapter 5, pp. 284-338. Intergovermental Panel for Climate Change, Cambridge, Cambridge University Press.

Brown, S., Sathaye, J., Cannel, M. & Kauppi, P. 1996. Management of forests for mitigation of greenhouse gas emissions. In R.T. Watson, M.C. Zinyowera & R.H. Moss, eds. Climate change 1995: impacts, adaptations, and mitigation of climate change: scientific-technical analyses, Chapter 24, pp. 775-797. Contribution of Working Group II to the Second Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge and New York, Cambridge University Press.

Cairns, M.A., Brown, S., Helmer, E.H. & Baumgardner, G.A. 1997. Root biomass allocation in the world’s upland forests. Oecologia 111: 1-11.

Detwiler, R.P. & Hall, C.A.S. 1988. Tropical forests and the global carbon cycle. Science 239: 42-47.

Dixon, R., Brown, S., Houghton, R.A., Solomon, A.M., Trexler, M.C. & Wisniewski, J. 1994. Carbon pools and flux of global forest ecosystems. Science 263: 185-190.

Friday, K.S., Drlling, M.E. & Garrity, D.P. 1999. Imperata grassland rehabilitation using agroforestry and assisted natural regeneration. Southeast Asian Regional Research Programme. Bogor, Indonesia, International Centre for Research in Agroforestry. 167 pp.

Frumhoff, P.C., Goetze, D.C. & Hardner, J.J. 1998. Linking solutions to climate change and biodiversity loss through the Kyoto Protocol’s Clean Development Mechanism. Cambridge, Massachusetts, USA, Union of Concerned Scientists. 14 pp.

Gintings, A.N. 2000. The progress of establishing cinnamon and candlenut demonstration plots in Jambi and their potential to absorb carbon. Paper presented in the Science and Policy Workshop on Terrestrial Carbon and Possible Trading under the CDM. 27 February-1 March 2000. IC-SEA, BIOTROP, Bogor, Indonesia.

Hairiah, K., von Noordwijk, M. & Palm, C. 1999. Methods for sampling above-and below-ground organic pools. In D. Murdiyarso, M. van Noordwijk & D.A. Suyamto, eds. IC-SEA Report No. 6: Modelling global change impacts on the soil environment, pp. 46-77. Bogor, Indonesia, BIOTROP-GCTE/Impacts Centre for Southeast Asia.

Hairiah, K. & Sitompul S.M. 2000. Assessment and simulation of above-ground and below-ground carbon dynamics. Bogor, Indonesia, APN/IC-SEA.

Houghton, R.A. 1996. Converting terrestrial ecosystem from sources to sinks of carbon. Ambio. 25(4): 267- 278.

Houghton, R.A. 1991. Releases of carbon to the atmosphere from degradation of forests in tropical Asia. Can. J. For. Res. 21: 132-142.

IPCC. 1996a. Impacts, adaptations and mitigation of climate change: scientific-technical analyses. Contribution of Working Group II to the Second Assessment Report of the Intergovernmental Panel on Climate Change. In R.T. Watson, M.C. Zinyowera & R.H. Moss, eds. Climate change 1995. Cambridge University Press.

IPCC. 1996b. Revised 1996 IPCC guidelines for national greenhouse gas inventories. Intergovernmental Panel on Climate Change, Geneva, Switzerland.

Iverson, R.I., Brown, S., Grainger, A., Prassad, A. & Liu, D. 1993. Carbon sequestration in tropical Asia: an assessment of technically suitable forest lands using geographic information systems analysis. Climate Research 3: 23-38.

Kawahara, T., Kanazawa, Y. & Sakurai, S. 1981. Biomass and net production of man-made forest in the Philippines. Jap. J. For. Soc. 63(a): 320-327.

Kummer, D.M. 1992. Deforestation in the post-war Philippines. Philippines, Ateneo de Manila University Press. 177 pp.

Lachica-Lustica, A. 1997. Trial planting of introduced reforestation species in Dumarao, Capiz. pp. 270-281. Developments in Agroforestry Research Book Series Number 160. ICRAF/FAO/APN/PCARRD/DOST.

Lasco, R.D. 2002. Forest carbon budgets in Southeast Asia following harvesting and land cover change. Science in China (Series C) 45: 55-64.

Lasco, R.D., Guillermo, I.Q., Cruz, R.V.O., Bantayan, N.C. & Pulhin, F.B. 2001. Carbon stocks assessment of a secondary forest in Makiling, Philippines. Intl. Tropical Forest Sci. J. (in press)

Lasco, R.D., Lales, J.S., Guillermo, I.Q. & Arnouevo, T. 1999. CO₂ absorption study of the Leyte geothermal forest reserve. Final Report of a study conducted for the Philippine National Oil Company. UPLB Foundation Inc.

Lasco, R.D.& Lasco, H.D. 1989. Agroforestry systems in the Philippines: inventory and assessment. Proc. Symposium on Agroforestry Systems and Technologies. 19-21 September 1989. BIOTROP, Bogor, Indonesia.

Lasco, R.D. & Pulhin, F.B. 2000. Forest land-use change in the Philippines and climate change mitigation. Mitigation and Adaptation to Climate Change Journal 5:81-97.

Lasco, R.D., Pulhin, F.B., Visco, R.G., Racelis, D.A., Guillermo, I.Q. & Sales, R.F. 2000. Carbon stocks assessment of Philippine forest ecosystems. Paper presented at the Science-Policy Workshop on Terrestrial Carbon Assessment for Possible Carbon Trading. 28-20 February 2000. Bogor, Indonesia.

Lasco, R.D. & Suson, P.D. 1999. A Leucaena leucocephala-based indigenous fallow system in central Philippines: the Naalad system. Intl. Tree Crops Journal 10(2): 161-174.

Lasco, R.D., Visco, R.G. & Pulhin, J.M. 2000. Formation and transformation of secondary forests in the Philippines. Paper used as background material for the International Workshop on Tropical Secondary Forests in Asia: Perspectives and Realities. April 2000. Samarinda, Indonesia, CIFOR.

Lugo, A.E. & Brown, S. 1992. Tropical forests as sinks of atmospheric carbon. Forest Ecology and Management 54: 239-255.

Lugo, A.E. & Brown, S. 1993. Management of tropical soils as sinks or sources of atmospheric carbon. Plant and Soil 149: 27-41.

Macandog, D.M. 2000a. Status of GHG inventory for the LULUCF sector in the Philippines, Indonesia and Thailand. Proceedings of the IGES/NIES Workshop on GHG Inventories for Asia-Pacific Region, pp. 110-137. Shonan Village, Japan.

Murdiyarso, D. 2000. Land-use change and terrestrial carbon stocks: capacity building, impacts assessment, and policy support in South and Southeast Asia. Final Activity Report. Bogor, Indonesia, Global Change Impacts Center for Southeast Asia.

Murdiyarso, D., Hairah, K., Husin, Yahya A. & Upik R. Wasrin. 1996. Greenhouse gas emissions and carbon balance in slash-and-burn practices. Workshop on Alternatives to Slash-and-Burn. Agroforestry Characterization in Jambi Povince, pp. 35-38.

Murdiyarso, D. & Wasrin, U.R. 1995. Estimating land-use change and carbon release from tropical forests conversion using remote sensing technique. J. of Biogeography 22: 715-721.

Noordwijk, M., Hairiah, K. & Sitompul, S.M. 2000. Reducing uncertainties in the assessment at national scale of C stock impacts of land-use change. In D.B. Macandog, ed. Proc. IGES/NIES Workshop on GHG Inventories for Asia-Pacific Region, pp. 150-163. Hayama, Japan, Institute for Global Environmental Strategies (IGES).

Pinard, M.A. & Putz, F.E. 1996. Retaining forest biomass by reduced logging damage. Biotropica 28(3): 278-295.

Pinard, M.A. & Putz, F.E. 1997. Monitoring carbon sequestration benefits associated with a reduced-impact logging project in Malaysia. Mitigation and Adaptation Strategies for Global Change 2: 203-215.

Prasetyo, L.B., Murdiyarso, D., Rosalina, U., Genya, S., Tsuruta, H., Okamoto, K., Shigehiro, I. & Shingo, U. 2000. Analysis of land-use changes and greenhouse gas emission (GHG) using geographical information system (GIS) technologies. Paper presented in the Workshop on Improving Land-use/cover Change and Greenhouse Gas Emissions Biophysical Data. 16 December 2000. Institute Pertanian Bogor, Indonesia.

Putz, F.E. & Pinard, M.A. 1993. Reduced-impact logging as a carbon-offset method. Conservation Biology 7(4): 755-758.

Schimmel, D., Enting, I.G., Heimann, M., Wigley, T.M.L., Rayneud, D., Alves, D. & Seigenthler, U. 1995. CO₂ and the carbon cycle. In J.T. Houghton, L.G. Meira Filho, J. Bruce, H. Lee, B.A. Callander, E. Haites, N. Harris & K. Maskell, eds. Climate change 1994: radiative forcing of climate change and an evaluation of the IPCC IS92 emision scenarios, pp. 35-71. Published for the Intergovernmental Panel for Climate Change, Cambridge University Press, Cambridge.

Siregar, C.A. & Gintings, A.N. 2000. Research activities related on ground biomass measurement at forestry research and development agency. Paper presented at the Workshop on Improving LUCC and Greenhouse Gas Emissions Biophysical Data. 16 December 2000. Institute Pertanian Bogor, Indonesia.

Siregar, S.T.H., Hardiyanto, E.B. & Gates, K. 1998. Acacia mangium plantations in PT Musi Hutan Persada, South Sumatra, Indonesia. In E.K.S. Nambiar, C. Cossalter, & A. Tiarks, eds. Site Management and Productivity in Tropical Plantation Forests, pp. 39-44. Bogor, Indonesia. Center for International Forestry Research.

Sitompul, S.M. & Hairiah, K. 2000. Biomass measurement of homegarden. Paper presented at the Workshop on Improving LUCC and Greenhouse Gas Emissions Biophysical Data. 16 December 2000. Institute Pertanian Bogor, Indonesia.

State Ministry for Environment. 1999. Indonesia: The first national communication under the United Nations Framework Convention on Climate Change, pp. 33-39. Jakarta, Indonesia.

Tjitrosemito, S. & Mawardi, I. 2000. Terrestrial carbon stock of oil palm plantation. Paper presented in the Science and Policy Workshop on Terrestrial Carbon and Possible Trading under the CDM. 27 February-1 March 2000. Bogor, Indonesia, IC-SEA, BIOTROP.

Trexler, M.C. & Haugen, C. 1994. Keeping it green: tropical forestry opportunities for mitigating climate change. Washington, D.C., World Resources Institute. 52 pp.

UNFCCC. 1997. The Kyoto Protocol. UNFCCC, Geneva, Switzerland.

Watson, R.T., Noble, I.R., Bolin, B., Ravindranath, N.H., Verado D.J. & Dokken, D.J. (eds). 2000. Land use, land-use change and forestry. Published for the Intergovernmental Panel for Climate Change. Cambridge, Cambridge University Press. 377 pp.

Weidelt, H.J. & Banaag, V.S. 1982. Aspects of management and silviculture of philippine dipterocarp forests. Eschborn, Germany, GTZ. 301 pp.

Zamora, D.S. 1999. Carbon dioxide (CO₂) storage potential of multistorey agroforestry sysytems in Mt. Makiling. College, Laguna, Philippines, College of Forestry and Natural Resources, UPLB. (MS thesis)