Forest management and agroforestry to sequester and conserve atmospheric carbon dioxide

P.E. Schroeder; R.K. Dixon and J.K. Winjum

Paul E. Schroeder is with ManTech Environmental Technology, Inc., at the USEPA Environmental Research Laboratory. Robert K. Dixon is with the USEPA Environmental Research Laboratory. Jack K. Winjum is with the National Council for Air and Stream Improvement at the USEPA Environmental Research Laboratory, Corvallis, USA.

An assessment of promising forest management practices and technologies, including site-level costs, for enhancing the conservation and sequestration of atmospheric carbon.

The accumulation of greenhouse gases in the atmosphere, particularly carbon dioxide (CO₂), is projected to alter the earth's climate. The potential role of forests in carbon sequestration has recently been evaluated by a number of authors (Marland, 1988; Andrasko, Heaton and Winnett, 199l; Grainger, 1991; Houghton, Unruh and Lefebvre, 1991; Sedjo and Solomon, 1991). Although they are preliminary, these analyses suggest that forest conservation, establishment and management as well as agroforestry could contribute to global carbon sequestration and conservation while providing goods and services in local communities of many countries. At the same time, the authors of these analyses agree on one critical point: forest carbon sequestration options alone will not solve the problems related to greenhouse gases. Addressing the climate change issue on a global scale will require complex adaptation and mitigating measures affecting all social and economic sectors. Moreover, it is clear that any forestry-based responses should represent a sound policy that is independent of the predicted global warming, and should produce net benefits in addition to those that may ultimately arise in the climate change context.

A forest plantation in the dry tropics of Latin America

In 1989, representatives from 67 countries at the Dutch Ministerial Conference held in Noordwijk, the Netherlands, discussed the role of world forests as carbon sinks. An agreement was reached to recommend to the Intergovernmental Panel on Climate Change that the goal of a net increase in world forest area of 12 million ha/year be adopted by the beginning of next century. This goal would be achieved through the conservation of existing forests, reforestation of degraded forest lands and afforestation of marginal, agricultural, pasture and savannah lands.

As part of the Global Change Research Program of the United States Environmental Protection Agency (USEPA), an assessment was initiated in 1990 to evaluate forest establishment and management options to sequester carbon and reduce the accumulation of greenhouse gases in the atmosphere. Three specific objectives were formulated:

· identify site-suitable technologies and practices that could be utilized to manage forests and agroforestry systems to sequester and conserve carbon;

· assess available data on site-level costs of promising forest and agroforestry management practices;

· evaluate estimates of technically suitable land in forested nations and biomes of the world to help meet the Noordwijk forestation targets.

This article summarizes the published report of the above-mentioned assessment (Dixon, Schroeder and Winjum, 1991).

A three-tiered agroforestry system in Costa Rica consisting of banana, coffee and agave

Approaches to the assessment

The first step in the USEPA assessment was the development of a global database, storing regional and national data from 94 nations in three major categories: forest growth and conservation resulting from forest establishment and management practices; the associated costs for each management practice; and the area of land potentially suitable for each practice. Information for the database was gathered from international scientific and technical literature published over the past ten years and from professional forestry contacts worldwide.

Although more than 90 nations are represented in the database, the assessment focuses on 16 key countries (Argentina, Australia, Brazil, Canada, China, Congo, Germany, India, Indonesia, Malaysia, Mexico, New Zealand, South Africa, United States, the former USSR and Zaire).

Carbon storage estimates

Data on forestry growth and yields provided the basis for estimates of potential carbon storage. Stem wood volume was converted to whole tree biomass, which was assumed to have a 50 percent carbon content. Whole tree biomass included roots, but not soil carbon, detritus or humus. Annual carbon accumulation and storage were derived by summing the carbon standing crop for each year in the growing cycle or rotation and dividing by the length of the rotation. This estimate of the average amount of carbon on site over a rotation applies to a series of rotations if we assume that the system is sustainable and there is no yield reduction in later rotations (Graham et al., 1990; Schroeder, 1992).

It is noteworthy that this approach also assumes that at harvest, or shortly after, most stored carbon returns to the atmosphere. This is based on an estimate that at least 40 percent of the total carbon in tree biomass is in leaves and branches, which are either burnt or decompose quickly after harvest. Of the remaining 60 percent of harvested carbon, in most processing operations, less than half of the total volume remains in final products. Therefore, more than 75 percent (and by some estimates as much as 90 percent) of stored carbon returns to the atmosphere quickly after harvest. This in no way negates the opportunities presented for long-term storage of carbon in finished wood-based products; rather, it emphasizes the need for more efficient harvesting and processing operations.

Cost estimates

Cost estimates were based on implementation costs per hectare for various forestry and agroforestry practices. Implementation cost reporting methods vary but generally include site preparation, stock costs and planting labour plus supervision. Thus, the estimates had three important shortcomings: the cost of land was not included because of obvious calculation difficulties; the analysis did not include annual or maintenance costs; neither social nor political variables affecting costs of management practices were considered. However, an analysis of annual maintenance costs and associated benefits is now being undertaken by USEPA.

Because forests are renewable resources, the costs of establishing them are recurring costs. The assessment used standard economic discounting techniques to compute the present value of a series of successive future costs. Published cost data were converted to US dollars and used to calculate the present value of all establishment costs for a 50-year period. The assumption of 50 years is arbitrary, but one that seems appropriate for a global programme of forest management that would require significant operational start-up time as well as a time-lag before planted trees reach their full potential (Grainger, 1991). Costs presented - see section, Cost of forest management options at site level, and Table 2 - are the sample medians, followed by an interquartile range (the middle 50 percent-of observations) in parentheses.

It is particularly important to emphasize that the costs presented are gross, i.e. they do not account for financial or social benefits that result from the initial investment or the production of useful products (Gregerson, Draper and Elz, 1989). To calculate net costs, the present value of future revenues as well as any subsidy inputs would have to be subtracted. As methods are developed and all benefits can be fully accounted for, it is likely that the value of the benefits will compensate for, if not exceed, a large portion of the costs. Three examples of analyses including both costs and benefits are provided in Table 1.

Analysis of promising management options

The conservation of world forests, particularly the reduction of deforestation in the tropics, is estimated to be one of the greatest potential contributions to offsetting the buildup of atmospheric CO₂. The causes of deforestation have been extensively reviewed; in order of importance, they broadly include shifting cultivation, the clearing of forests for animal grazing and exploitive tree harvests. The replacement of shifting agricultural practices with agroforestry could potentially reduce deforestation (Wiersum, 1990; MacDicken and Vergara, 1990) and, consequently, reduce carbon emissions from tropical regions.

Direct evidence of this potential is limited, but one research study indicated that a low input agroforestry system, involving the rotation of acid-tolerant crops, produced agricultural products on a single hectare equivalent to the volume normally produced on 5 to 10 ha under slash-and-burn agriculture (Sanchez and Benites, 1987). One hectare of closed tropical forest can contain up to 220 tonnes of carbon (t C) most of which, when burnt, is released into the atmosphere (Waring and Schlesinger, 1985). Therefore, for each hectare of agroforestry established on deforested land in the tropics, perhaps as much as 2200 t C could be prevented from going into the atmosphere. Of course, agroforestry occurs in many forms which may be more or less productive.

TABLE 1. Costs and benefits of three sustainable agricultural systems in Latin America

	Extensive agroforestry	Low-input cropping	Intensive agroforestry
	(US$)
Costs (labour, materials)/ha	47	737	767
Gross revenues/ha	76	2229	1059
Net revenues/ha	29	1492	292
General assumptions
Agricultural extension costs/ha	5	5	5
	(hectares)
Deforestation avoided/year	5	4.6	20
	(US$)
Reforestation costs avoided/year	3500	3220	14000
	(tonnes)
Carbon emission from deforestation avoided/year	350	322	1400
	(US$)
Costs and benefits
Total costs/ha	52	742	772
Total benefits/ha	3529	4712	14292

Source: Andrasko, Heaton and Winnett, 1991.

Regarding the expansion of world forests, an analysis of the database information on potential carbon sequestration values according to different management practices helps to determine which are most promising. The five most promising practices in terms of median value for carbon sequestration (in tonnes per hectare) are discussed below in descending order.

Natural regeneration in the tropics. This practice has a median value of 195 t C/ha over a period of 50 years. This is the highest median value found in the database and it possibly reflects the elevated biomass productivity rates of natural ecosystems in the humid tropics. However, this practice currently has only three references in the database and needs more verification.

Afforestation in temperate latitudes. The median value is 120 t C/ha, a high value which probably reflects the high growth rates of plantations established on lands removed from agricultural production. Although these lands may have been only marginally productive in terms of agricultural crops, for forest plantations they have often represented particularly high-potential sites (Hughes, 1991).

Agroforestry in the tropics. The assessment shows a median value for this practice of 95 t C/ha. Tropical agroforestry has been under intensive study for only a decade (MacDicken and Vergara, 1990) and, in comparison with reforestation, there are consequently fewer data while those that are available may be skewed towards high-end results. Nonetheless, the values for agroforestry are encouraging because this practice will certainly be important from the standpoint of supporting local populations.

Reforestation in the tropics. This practice has a median carbon sequestration value of 65 t C/ha. The 136 entries in the database lend considerable weight to the validity of these sequestration values while the high ranking of this practice supports the conclusion, often advanced, that reforestation in the tropical latitudes has great potential.

Reforestation in the temperate latitudes. At a median value of 56 t C/ha, this approach is the fifth highest on the list of promising practices for carbon sequestration. The estimates are based on 212 entries, the largest number in the database. Reforestation, as compared to "afforestation", described above, usually involves less dense stands on poorer soils, thus accounting for the difference in sequestration rates.

It is noteworthy that silvicultural practices had the lowest median value among management practices in all three latitude regions (boreal = 10 t C/ha; temperate = 26 t C/ha; tropical = 34 t C/ha). Silvicultural treatments, such as thinning and fertilization in plantations, will almost certainly play an important role in adapting forests to the predicted changes in climate. However, by themselves, such treatments will not produce an increase in forest area and their potential to offset the buildup of atmospheric CO₂ appears to be low.

TABLE 2. Cost efficiency of selected forestry practices in different latitudes

Zone/forestry practice	Median	Median
	($/t C)	($/ha)
Boreal
Natural regeneration	5	93
	(4-11)	(83-126)
Reforestation	8	324
	(3-27)	(127-455)
Temperate
Natural regeneration	1	9
	(<1-1)	(9-10)
Afforestation	2	259
	(<1-5)	(41-444)
Reforestation	6	357
	(3-29)	(257-911)
Tropical
Natural regeneration	1	178
	(<1-2)	(106-238)
Agroforestry	5	454
	(2-11)	(255-699)
Reforestation	7	450
	(3-26)	(303-1183)

Note: Sample medians are followed by an interquartile range of observations in parentheses.

Cost of forest management options at site level

The site-level cost of implementing promising carbon sequestration options was one of the primary objectives of the overall assessment. Implementation or initial costs of forest establishment and management generally appear to be lowest in boreal regions. As management intensity increases in temperate and tropical regions, initial costs per hectare escalate accordingly (Dixon, Schroeder and Winjum, 1991).

For the boreal forest system, natural regeneration and artificial reforestation could sequester carbon most efficiently at a cost of US$93 to $324/ha. At sequestration values of about 17 t C/ha and 39 t C/ha for a 50-year period, the initial costs for the two practices are $5 ($4 to $11) and $8 ($3 to $27)/t C, respectively.

In temperate regions, reforestation, afforestation, natural regeneration and silvicultural practices offer the least expensive opportunities for sequestering carbon. Artificial reforestation can cost $357/ha at a sequestration rate of 56 t C/ha. Carbon is stored at an initial cost of $6 ($3 to $29)/t C depending on site conditions, tree species and management intensity. Afforestation can store about 120 t C/ha at a cost of $259/ha or $2 ($0.22 to $5)/t C. Natural regeneration can be very inexpensive at less than $$10/ha or at 9 t C/ha, the cost is less than $1 ($0.01 to $0.43)/t C.

The widest range of costs was reported for forest carbon conservation or sequestration options in tropical latitudes. Natural regeneration, short-rotation fuelwood plantations and agroforestry systems can all be established for less than $1000/ha (50-year cost basis). Reforestation and agroforestry can sequester carbon at less than $10 ($2 to $26)1t C because of the high sequestration values. Intermediate silvicultural treatments stimulate productivity and can sequester carbon at approximately $500/ha or $8.50 ($1.50 to $36)/t C at a sequestration value of 59 t C/ha (see Table 2 for cost efficiency of forestry practices in different latitudes).

Intensive short rotation plantation of radiata pine in New Zealand

What is the overall potential and cost?

A marginal cost analysis integrated data on carbon storage, establishment costs and land area. Forest management practices and their associated potential land areas were ranked in ascending order from the least to the most expensive per tonne of stored carbon. Assuming that it is most rational to begin by storing the least expensive carbon (Moulton and Richards, 1990), we can accumulate both carbon and costs as we go through the ranked list, adding ever more expensive practices. This approach showed that the marginal cost of storing 45 to 65 gross tonnes (Gt) of C (1 Gt = 10⁹ t) would be about $3/t C with a total cost of $135000 million to $195000 million. At more than 70 Gt C the marginal cost escalated sharply to over $100/t C. Storing 45 to 65 Gt C would require from 400 to 950 million ha.

For a clearer perspective on these cost estimates, it is noteworthy that a recent study by the United States National Academy of Sciences categorized greenhouse gas mitigation options that cost less than $33/t C as low-cost (NAS, 1991). A policy of taxing carbon emissions from fossil fuels could cost as much as $100/t C (OTA, 1991). A subsidy of $0.02/kwh to promote the use of non-fossil fuel energy sources could cost from $75 to $150/t C.

Land suitability

The picture of land suitability and availability is clouded by a lack of reliable data and complicated by economic, social and land-use issues. Nonetheless, there appears to be a large area of land in the world that is available for and would benefit from tree-planting. For the tropics alone, Grainger (1991) estimated that there are 621 million ha which are technically suitable for the establishment of tree plantations; Houghton, Unruh and Lefebvre (1991) estimated that there are 579 million ha available for plantation establishment, 858 million ha for natural regeneration and regrowth and 500 million ha for agroforestry. Trexler (1991) attempted to factor in social and competing land-use constraints and estimated that, for tropical Africa and Asia, there are 46 million ha available for plantation establishment, 163 million ha for natural regeneration and 102 million ha for agroforestry.

Approximate areas of suitable land in the boreal regions appear to be 100 million to 200 million ha (Volz, Kriebitzsch and Schneider, 1991); and in temperate regions 200 million to 300 million ha (Moulton and Richards, 1990; Volz, Kriebitzsch and Schneider, 1991).

A conceptual approach to the Noordwijk target

Can forest management and agroforestry practices throughout the world undergo large-scale expansion soon enough to be of significant aid in offsetting the buildup of atmospheric CO₂? Using the forest goals of the Noordwijk Declaration as a framework, the stepwise approach presented below suggests a positive answer to this key question.

The Noordwijk proposal is to achieve a net annual increase of 12 million ha of new forest area over world deforestation. Based on current estimates of deforestation of about 17 million ha/year (Allen and Lanly, 1991) and projections that this rate could reach 30 million ha/ year by 2045 (Houghton, 1990, Myers, 1986), implementation of the Noordwijk. Declaration would require more intensive forest practices on about 30 million to 40 million new ha/year. The following discussion assumes a target of 35 million ha/year for the period 2000 to 2040.

An "easy first" paradigm

With regard to a global net forestation plan, an "easy first" paradigm is suggested. Using this approach, plans make allowance for the programme to start where the obstacles are minimal. Simultaneously, research and negotiations can be undertaken to resolve obstacles. The "easy first" approach is not suggested as a total solution. It is overly simple in relation to the many day-to-day resource constraints as well as the specific social, economic and political issues within individual nations. But the point that the "easy first" approach attempts to illustrate is that all financial commitments, socio-political agreements, technical expertise, etc., although they are critical, do not have to be fully in place before work can start.

In an analysis of forest management practices with potential to slow deforestation and increase forest area, Andrasko (1990) suggests three strategies: i) maintain forest area; ii) reduce loss of forests; iii) expand forest area. Following are some possible approaches to the 35 million ha/year target using this scheme.

Agriculture plus forest conservation. As a start, increasing sustainable agricultural practices in the tropics would contribute significantly to strategies i) and ii). Sanchez (1990) estimates that for every hectare placed in sustainable soil management for agriculture, 5 to 10 ha (assuming an average of 7.5 ha) of tropical forest could be conserved. Ross-Sheriff and Cough (1990) suggest that it is not unreasonable that a rate of about 1 million ha/year could be achieved in ten years, starting with a first-year level of 50000 ha. If the ratio of sustainably managed hectares to reduced deforestation averages 7.5:1, then 7.5 million ha of the 35 million ha goal could be achieved by the year 2000. Of course, the actual outcome will also depend on the future path of population growth, but fivefold to tenfold increases in agricultural output would have a major impact.

Other approaches cited by Andrasko (1990) for strategies i) and ii) are forest reserves; extractive reserves; natural forest management; and the increased use of pastures (increasing carrying capacity through improved management practices, including seeding and fertilization, silvipastoral practices, reclamation of degraded pastures, etc.).

Assuming that the potential of these approaches could offset another 2.5 million ha/year of deforestation by the turn of the century, perhaps 10 million ha of the 35 million ha goal could be achieved by these practices.

Agroforestry. For agroforestry, Houghton, Unruh and Lefebvre (1991) estimate that about 500 million ha of former forest land might be available (60 million ha of degraded forest land; 38 million ha of woodland; 402 million ha of grassland).

If implementation of agroforestry could increase to an annual rate of 1 percent of the total, then as much as 5 million ha of new tree-growing land could be added each year.

Reforestation and afforestation. To increase forest area, both reforestation and afforestation are required at increasing rates. In this discussion, it is assumed that the annual rates of reforestation and afforestation of marginal lands can be increased by the year 2000 to 2 million ha/year in the boreal latitudes and to 3 million ha/year in both the temperate and tropical latitudes.

Restoration of degraded lands. The tropical latitudes have the primary potential for restoring degraded lands. Grainger (199l) estimates that 621 million ha of degraded tropical land have the potential to support forest plantations. Houghton, Unruh and Lefebvre (1991) conclude that about 580 million ha of degraded but ecologically suitable lands could he put into plantations. Assuming, therefore, that 600 million ha is a reasonable approximation of the amount of degraded land in the tropics today, the suggested Noordwijk goal for restoration is about 10 million ha/year starting in 2000. In India, approximately 1 million ha of degraded land is reclaimed each year (Sharma, Sharma and Garcha, 1989). For the boreal and temperate latitudes, it is also assumed that each of these areas would have approximately 1 million ha of degraded land that could be restored annually.

TABLE 3a. Forest management in the former USSR: a boreal example

Forest options	Availability for expansion				"Easy first" ranking	Proportion of goal	Priority for *R & N
	Capital	Labour	Land	Technology	(million ha)
Maintaining forest area
a) Protection of forest reserves	Required by all nations at current levels
b) Extractive reserves	No	Moderate	No	Moderate	Low		Moderate
Reducing loss of forests
a) Natural forest management	Moderate	Yes	Yes	Yes	High	1.000	Low
b) Increased use of pastures	Moderate	Moderate	No	Moderate	Low		Moderate
c) Sustainable agriculture	Moderate	Yes	Yes	Moderate	Moderate	0.525	Low
d) Agroforestry	No	No	No	No	Low		Moderate
Expanding forest area
a) Reforestation (and afforestation)	Yes	Yes	Yes	Yes	Very high	3.500	Low
b) Restoration of degraded lands	Moderate	Yes	Moderate	Moderate	Moderate	3.000	High
Total						8.025

Note: Total present forested area: 929 million ha; potential additional forested area (15%): 139 million ha: current harvest rate: 3.1 million ha; Current reforestation rate 4.5 million ha; Noordwijk Declaration goal for the former USSR: 8 million ha.
* R & N = research and negotiations.

TABLE 3b. Forest management in the United States: a temperate example

Forest options	Availability for expansion				"Easy first" ranking	Proportion of goal	Priority for *R & N
	Capital	Labour	Land	Technology	(million ha)
Maintaining forest area
a) Protection of forest reserves	Required by all nations at current levels
b) Extractive reserves	Moderate	Moderate	Moderate	Moderate	Low	0.025	Moderate
Reducing loss of forests
a) Natural forest management	Moderate	Moderate	Yes	Moderate	Low	0.300	Moderate
b) Increased use of pastures	Yes	Yes	Moderate	Yes	Moderate	0.250	Moderate
c) Sustainable agriculture	Yes	Yes	Yes	Yes	Very high	0.500	Low
d) Agroforestry	Yes	Yes	Moderate	Yes	High	0.250	Moderate
Expanding forest area
a) Reforestation (and afforestation)	Yes	Yes	Moderate	Yes	High	0.750	Moderate
b) Restoration of degraded lands	Moderate	Moderate	Yes	Yes	Moderate	0.500	High
Total						2.575

Note: Total present forested area: 298 million ha; potential additional forested area (15%): 44.7 million ha current harvest rare: 2 million ha; current reforestation rate: 1.8 million ha Noordwijk Declaration goal for the United States: 2.6 million ha.
* R & N - research and negotiations.

TABLE 3c. Forest management in Brazil: a tropical example

Forest options	Availability for expansion				"Easy first" ranking	Proportion of goal	Priority for *R & N
	Capital	Labour	Land	Technology	(million ha)
Maintaining forest area
a) Protection of forest reserves	Required by all nations at current levels
b) Extractive reserves	Moderate	Yes	Yes	Moderate	Moderate	0.200	High
Reducing loss of forests
a) Natural forest management	Moderate	Yes	Yes	Yes	High	0.500	High
b) Increased use of pastures	Moderate	Yes	Yes	Moderate	Moderate	0.250	Moderate
c) Sustainable agriculture	Moderate	Yes	Yes	Yes	High	1.000	Moderate
d) Agroforestry	Yes	Yes	Yes	Yes	Very high	1.000	Moderate
Expanding forest area
a) Reforestation (and afforestation)	Moderate	Yes	Yes	Yes	High	0.750	Moderate
b) Restoration of degraded lands	Moderate	Yes	Yes	Low	Moderate	0.750	High
Total						4.450

Note: Total present forested area: 514.5 million ha; potential additional forested area (15%):77.2 million ha; current harvest rate 34 million ha; Current reforestation rate: 0.45 million ha. Noordwijk Declaration goal for Brazil: 4.5 million ha.
* R & N = research and negotiations.

In total, therefore, these estimates reflect a large enough potential pool of land for expanding the world's present forest area by about 20 million ha annually. Reforestation and afforestation would account for 8 million ha and the restoration of degraded land for 12 million ha.

National forestation goals

Assuming an international commitment to achieving the Noordwijk goal, nations with forest resources would then need to select from a list of forest management options and make their contribution to the overall target according to the size of their respective forest land base while, most importantly, adopting the "easy first" concept within the framework of an international agreement designed to ensure equity for all nations. Possible national goals for three areas (the former USSR, the United States and Brazil) are presented as examples for boreal, temperate and tropical latitudes, respectively, in Table 3a,b,c.

Limitations and needs for future research

The current assessment of biological and cost information from more than 90 countries worldwide represents the first significant attempt to develop a bottom-up global analysis. However, before practices can be widely and successfully implemented, consideration must be given to the array of possible economic and socio-political constraints.

A key concern is land availability. Data on the suitability of land for reforestation and other forest management practices are highly variable (compared with data on carbon storage and costs) and this is a serious constraint. It is also important to draw a distinction between land that is technically suitable for reforestation and forest management and land that is actually available (Grainger, 1991; Trexler, 1991; Winjum, Schroeder and Kennedy, 1991). Because of population pressures and demands from competing land uses, much of the land that appears to be technically suitable for forest management may not be available.

Moreover, the degree to which present day forest management practices can simultaneously accomplish objectives as different as the maintenance of global forest area, sustainable economic development and the conservation of biodiversity, among others, is still untested. Research by USEPA and other national and international groups is recommended on the following topics: biomass productivity; costs of the full range of forest management practices; benefits of forest management practices; cost-benefit analysis; risk, uncertainty and constraints in forest and tree establishment and management; land use and tenure; and related issues.

A hardwood boreal forest in Russia

Conclusion

Overall, the biological opportunity to conserve and sequester carbon in the terrestrial biosphere, especially in forest systems, appears significant. With careful planning and implementation, management practices useful for this carbon benefit would appear to have potential to provide food, water, wood and other basic human needs. Although implementation costs seem modest, a primary research objective is to place reliable values on all possible forest benefits. Benefit values would allow a clearer determination of the net costs of sequestering carbon through forest management and agroforestry systems. On a global basis, these benefit values, together with more accurate estimates of land availability, will ultimately lead to a more definitive assessment of promising forest and agroforestry practices for sequestering and conserving carbon.

Bibliography

Allan, T. & Lanly, J.P. 1991. Overview of status and trends of world forests. In D. Howlett & C. Sargent, eds. Proc. Tech. Workshop to Explore Options for Global Forest Management, April 1991. Bangkok, Thailand. London, IIED.

Andrasko, K. 1990. Global warming and forests: an overview of current knowledge. Unasylva, 41 (163): 3-11.

Andrasko, K., Heaton, K. & Winnett, S. 1991. Estimating the costs of forest sector management options: overview of site, national and global analyses. In D. Howlett & C. Sargent, eds. Proc. Tech. Workshop to Explore Options for Global Forest Management, April 1991, Bangkok, Thailand. London, IIED.

Dixon, R.K., Schroeder, P.E. & Winjum, P. 1991. Assessment of promising forest management practices and technologies for enhancing the conservation and sequestration of atmospheric carbon, and their costs at the site level. EPA/600/3-91/067. Corvallis, Oregon, USEPA.

Graham, R.L., Perlack, RD., Prasad, A.M.G., Ramney, J.W. & Waddle, D.B. 1990. Greenhouse gas emissions in sub-Saharan Africa. No. ORNL-6640. Tennessee, United States Department of Energy, Oak Ridge National Laboratory.

Grainger, A. 1991. Overcoming constraints on assessing feasible afforestation and revegetation rates to combat global climate change. In D. Howlett & C. Sargent, eds. Proc. Tech. Workshop to Explore Options for Global Forest Management. April 1991. Bangkok. Thailand. London, IIED.

Gregerson, H., Draper, S. & Elz, D. 1989. People and trees: the role of social forestry in sustainable development. EDI seminar series. Washington. DC. World Bank.

Houghton, R.A. 1990. The future role of tropical forests in affecting the carbon dioxide concentration of the atmosphere. Ambio, 19(4): 204-209.

Houghton, R.A, Unruh, J. & Lefebvre, P.A. 1991. Current land use in the tropics and its potential for sequestering carbon. In D. Howlett & C. Sargent, eds. Proc. Tech. Workshop to Explore Options for Global Forest Management, April 1991, Bangkok, Thailand. London, IIED.

Hughes, J.H. 1991. A brief history of forest management in the American south: implications for large-scale reforestation to slow global warming. In J.K. Winjum, P.E. Schroeder &: M.J. Kennedy, eds. Proc. Int. Workshop on Large-scale Reforestation, EPA/600/9-91/014. Washington, DC, USEPA.

MacDicken, K.G. & Vergara, N.T. 1990. Agroforestry: classification and management. New York, Wiley.

Marland, G. 1988. The prospect of solving the CO₂ problem though global reforestation. DOE/NBB-0082. Washington, DC, United States Department of Energy, Office of Energy Research.

Moulton, RJ. & Richards, K.R. 1990. Costs of sequestering carbon through tree planting and forest management in the United States. United States Department of Agriculture Forest Service General Technical Report WO-58. Washington, DC, USDA Forest Service.

Myers, N. 1986. Tropical forests and their species: going, going...? In E.O. Wilson, ed. Biodiversity. Washington, DC, National Academy Press.

NAS. 1991. Policy implications of greenhouse warming. United States National Academy of Science. Washington, DC, National Academy Press.

Noordwijk Ministerial Conference. 1989. The Noordwijk Declaration on Climate Change. Ministerial Conf. Atmospheric Pollution and Climate Change, Noordwijk. The Netherlands, Government of the Netherlands.

OTA. 1991. Changing by degrees: steps to reduce greenhouse gases. Washington, DC, Office of Technology Assessment, Congress of the United States.

Ross-Sheriff, B. & Cough, P. 1990. Tropical forestry response options workshop. In Proc. Conf. Tropical Forestry Response Options to Global Climate Change, January 1990, São Paulo, Brazil. Washington, DC, Office of Policy Analysis, USEPA.

Sanchez, P.A. 1990. Deforestation reduction initiative: an imperative for world substantiality in the twenty-first century. In A. F. Bouwman, ed. Soils and the greenhouse effect. New York, Wiley.

Sanchez, P.A & Benites, J.R. 1987. Low-input cropping for acid soils of the humid tropics. Science, 238: 1521-1527.

Schroeder, P.E. 1992. Carbon storage potential of short rotation tropical tree plantations. Forest Ecology and Management, 50: 31-41.

Sedjo, R.A. & Solomon, A.J. 1991. Climate and forests. In N.S. Rosenberg, W.E. Easterling, P.R. Crosson & J. Dormstadter, eds. Greenhouse warming: abatement and adaptation. Workshop proceedings. Washington, DC, WRI.

Sharma, R.N., Sharma H.L. & Garcha, H.S. 1989. Fuelwood from wasteland. New Delhi, Bioenergy Society of India.

Trexler, M.C. 1991. Estimating tropical biomass futures: a tentative scenario. In D. Howlett & C. Sargent, eds. Proc. Technical Workshop to Explore Options for Global Forest Management, April 1991, Bangkok, Thailand. London, IIED.

Volz, H., Kriebitzsch, W. U. & Schneider, T.W. 1991. Assessment of potential, feasibility and costs of forestry options in temperate and boreal zones. In D. Howlett & C. Sargent, eds. Proc. Technical Workshop to Explore Options for Global Forest Management, April 1991, Bangkok, Thailand. London, IIED.

Waring, R.H. & Schlesinger, W.H. 1985. Forest ecosystems: concepts and management. Orlando, Florida, Academic Press.

Wiersum, K,F. 1990. Planning agroforestry for sustainable land use. In W.W. Budd, I. Duchart, L.H. Hardesty & F. Steiner, eds. Planning for agroforestry. Amsterdam, Elsevier.

Winjum, J.K., Schroeder, P.E. & Kennedy, MJ., eds. 1991. Proc. Int. Workshop on Large-scale Reforestation. EPA/600/9-91/041. Washington, DC, USEPA.