Common problems and opportunities

The previous chapter provides a comprehensive overview of (a) the important products, functions and services that forestry and agroforestry contribute-both directly and indirectly-to food security in the Asia-Pacific region; and (b) the key issues and factors that influence the linkages between forestry/agroforestry and food security.

From this overview, common problems and opportunities emerge for enhancing the contributions of forestry and agroforestry in the region. This chapter focuses on the common constraints and policy status of LIFDCs. It also highlights major problems, opportunities and success stories of how forestry and agroforestry can increase food security in certain problematic settings: major watersheds and forested areas, dry zones, temperate zones, small islands, and urban areas.

As an attempt to produce a preliminary typology, the following table indicates the political/economic groupings and the predominance of certain areas that are problematic for food security in the 19 countries in Asia-Pacific classified as LIFDCs (as of March 1996).

Table 1. Typology of LIFDCs in Asia-Pacific.

^{ASEAN: Association of Southeast Asian Nations; EIT: Economies in Transition;
MRC: Mekong River Commission; SAARC: South Asian Association for Regional Cooperation
SPF: South Pacific Forum}

A. Common constraints

1. Institutional

Among the most common constraints to improving the linkages between forestry/agroforestry and food security are institutional barriers in LIFDCs, as well as other countries in the region. Forestry and agricultural institutions generally have different organizational mandates and territoriality, which restrict information flow and interagency collaboration (Nair et al. 1995). Arbitrary grouping of administrative functions under different departments and ministries further accentuate structural barriers, especially in agroforestry and other interdisciplinary fields. These barriers are reinforced by three major developments (Lai and French 1994):

• The rather arbitrary classification of "forest" and "agriculture" land based solely on slope. Land above 18% slope in the Philippines is considered "uplands"; land steeper than 45% in Indonesia and Thailand is classified as "forest" land. This has placed most sloping upland areas (even those with good agricultural suitability) under forest department jurisdiction, and lowland areas under agricultural agencies.
• The trend towards separate ministries and departments for agriculture and forestry. In the past, most Asian countries grouped agriculture and forestry within the same ministry. In recent years, however, many countries have established separate forestry and/or environment ministries (e.g., Bangladesh), or have separated forestry research from the agricultural research (e.g., Indonesia).
• The separation of education and training systems for agriculture and forestry. Separate universities, faculties or academies reinforce the disciplinary barriers between agriculture and forestry, and hinder the development of interdisciplinary programmes such as agroforestry. Also, in some countries, forestry colleges or schools provide in-service training after foresters are recruited, while agricultural training is provided by universities and other institutions.

Institutional barriers between forestry and agriculture hamper the development of the crucial intersectoral linkages that could substantially contribute to food security. Tropical Forestry Action Plans/National Forestry Action Programmes and Forestry Sector Master Plans have been developed in many countries throughout the world, with support from donor and international organizations One criticism about such past efforts is that they did not adequately examine the contributions and linkages of the forestry sector to sustainable agricultural production (WRI 1990, TFAP 1990). Hence, there were many missed opportunities to generate better policy and more effective planning support for intersectoral development.

2. Production

Forestry and agroforestry production constraints relate closely to the available (and in many cases, shrinking) natural resource base. With the increasing trends of population growth and land degradation/conversion in Asia-Pacific, pressures are mounting on the dwindling forest and agroforestry resource base.

There are basic trade-offs related to the conversion of forest areas for agricultural purposes. In the past, many lowland forests were converted to rice-growing areas, which were essential to support the high population densities in fertile delta regions. The results of transmigration and resettlement programmes in Indonesia and other countries have been mixed. In many forest areas cleared for resettled families, the fragile tropical soils could not sustain agricultural production. To improve national food security, the Indonesian government plans to convert some 1.5 million ha of low-lying peat forest lands in Kalimantan and Sumatra. With all forest conversion activities, there will be at least short-term increases in food production, but the long-term benefits-both direct and indirect-of forests to food security will be lost. These trade-offs must be fully examined to determine whether a given conversion activity will actually contribute to long-term food security.

Of the approximately 6 billion people on earth, more than half live in Asia-Pacific. At the same time, the annual deforestation rate in Asia-Pacific is the highest in the world at 1.2% (or 3.9 million ha year) when compared to the global deforestation rate of 0.8% per annum (FAO 1993a). The result is a very low forest per capita ratio in the region-0.19 ha/person as opposed to the world average of 0.64 ha/person (FAO 1995a). Clearing of land for agricultural purposes is the leading cause of deforestation. The area of new lands being cultivated in the world each year is roughly equal to the total area deforested annually-about 15 million ha (Durst 1995). Other major causes of deforestation include: commercial logging, wood extraction and fisheries development in mangrove forests, and forest conversion due to urbanization and industrialization. In the years to come, this last factor is likely to be more dominant as many countries in Asia-Pacific accelerate their rates of economic development.

Deforestation has two major effects on food security. First, the direct production of forest foods-fruits, nuts, leaves, roots, mushrooms, saps, gums, starches, wildlife, insects and many other types-is lost on about 4 million ha of deforested land in Asia-Pacific each year. This adversely affects forest-dwelling and forest-dependent people who rely on forest foods for an important portion of their food requirements as supplements, seasonal and emergency substitutes, delicacies, or even as staples. For example, the sago palm, Metroxylon spp., is the main energy food for at least 300,000 people in Melanesia and as part of the diet of over 1 million people (Ulijaszek 1983). As recently as 1988, people living in wooded areas of northeastern Thailand obtained up to 60% of their food directly from the forests (FAO 1988a).

Secondly, deforestation erodes the positive, indirect influences of forests on food security-protecting watersheds, reducing soil erosion, improving soil fertility, producing livestock fodder, and providing silvofishery habitats (FAO 1989; see also sub-chapter 1.A). While these types of indirect losses in food production are difficult to quantify, they can be dramatic and even life-threatening. Examples include the devastating landslides in southern Thailand in 1988 and in Leyte, Philippines in 1990, which killed thousands of people and were largely due to inappropriate conversion of natural forests to rubber and sugar plantations (Brooks 1993, Durst 1994).

The phenomenon of "agrodeforestation" also occurs widely within the region. Originally coined in the Pacific Islands context, agrodeforestation refers to "the removal of trees or the de-emphasis on the planting and/or protection of trees in the context of existing agroecosystems." (Thaman 1996). In this light, the process of agrodeforestation is proceeding rapidly in many LIFDCs in Asia and the Pacific due to several factors, including: land acquisition and development, agricultural mechanization and marketization, and promotion of export commodity crops. These factors tend to favor the development of large-scale monocultural operations at the expense of small-scale, traditional agroforestry systems. This reduces the wide range of food products from agroforestry systems, and the biodiversity and environmental functions provided by these systems.

3. Technology

Despite considerable investment in forestry and agroforestry research, there remains many technological constraints to forestry and agroforestry development in support of food security. These constraints include:

• Germplasm availability: lack of readily available, high-quality planting materials for trees and other agroforestry species.
• Technology generation: researchers dominate technology generation process; linkages often weak with extension workers, farmers and other client groups; indigenous and farmer-generated knowledge not well utilized.
• Technology transfer: inaccessibility of technical information among farmers and grassroots organizations; existing extension organizations often promote inappropriate technologies; top-down approaches generally less effective than farmer-to-farmer mechanisms.

The selection of appropriate technologies is critical to the success or failure of forestry and agroforestry interventions. For example, an agroforestry system may use:

• Indigenous (existing) technologies that farmers are familiar with.
• Practices that have been modified or improved by farmers or outsiders.
• Practices introduced by outside researchers or extension agents.

Indigenous knowledge and existing practices should be the basis for designing agroforestry interventions. But many projects have relied (largely unsuccessfully) on introducing agroforestry models using exotic tree species and technologies. Introducing technologies from outside has two major risks:

• Introduced technologies may not be socially, culturally or economically acceptable to farmers.
• Introduced technologies or species may not be ecologically sound (e.g., introduction of Leucaena on acid soils).

Small-scale, on-farm experiments to test new technologies must be properly conducted and evaluated before these technologies can be promoted (IIRR and FAO 1995).

Swaminathan (1996a) advocates an "ecotechnological" approach to ensure sustained availability of food. Ecotechnology involves the blending of the ecological prudence and technologies of the past, with the best in frontier technologies related to biotechnology, information, space, renewable energy and management.

4. Access and equity

Access and equity issues are the most serious constraints to attaining food security in Asia-Pacific, much more so than production factors.

The 23rd FAO Regional Conference, held in Samoa in May 1996, crafted a "Asia-Pacific Perspective on Food Security," which noted, inter alia, that: "The aggregate food supply for the region as a whole is not the greatest problem; the problem now and in the near future is its inequitable distribution at the country and household levels."

The FAO-sponsored World Food Summit held in November in Rome reaffirmed the "right of everyone to have access to safe and nutritious food." It is ironic to note that the United States, the world's largest single food aid provider, took issue with this reaffirmation of the "right to food" fearing that it could lead to lawsuits from poor nations seeking aid and special trade provisions.

Forestry and agroforestry can directly provide food, ensure adequate nutrition, contain sources of seasonal or famine foods, and sustain agricultural production. Community forestry and agroforestry initiatives often target resource-poor farmers who are most vulnerable to food shortages and malnutrition. However, there are serious access and equity problems to overcome. Many of these problems are related to policies that restrict access to forestry and agroforestry resources. For example, in government-controlled forest areas,there are often prohibitions or restrictions on the collection of forest foods and products, and the fuelwood needed to convert food supplies to adequate diets.

In the allocation of state forest lands for agroforestry development-such as in Bangladesh and Indonesia-the powerful local elite may co-opt the process and obtain plots intended for allocation to marginal or landless farmers.

Hence, while forestry and agroforestry hold great potential for improving assess and equity to food sources, secure tenure and use rights must be provided to realize this potential. This will require some policy reorientation, particularly with restrictive forestry codes designed to keep people out of the forests.

5. Trade

Trade issues, particularly those that affect food security, are among the most contentious, hotly-debated issues between wealthy, food-exporting nations and LIFDCs. Depending on the view taken, "free trade" may either be the savior or destroyer of food security in developing countries. International trade encompasses many complex and emotional facets which are usually inseparable from sovereignty concerns and vested interests.

The United States led the group of rich nations at the World Food Summit in calling to make free trade the key to ending poverty as the root of hunger. However, critics counter that there is no free market in world agriculture. The industrialized countries annually spend US$185 billion to subsidize the production and export of farm surpluses, a sum equal to 40% of the value of production. Each farmer in the U.S. receives an average of about US$16,000 per year in subsidies to produce surpluses for export. "Subsidized dumping" of cheap food in developing countries can undermine local agriculture, rural development and create a dependence of foods like wheat, which cannot be adequately produced locally. In the Philippines, the government estimates that almost half-a-million farming livelihoods will be lost over the next decade as a result of "free trade" in food and agriculture (Watkins 1996).

Beyond these contentious macro-issues, many trade and marketing arrangements among LIFDCs and other countries can be beneficial for food security as well as national and household economies. Enormous potential exists for community-based production and marketing of non-wood forest products and agroforestry products for both domestic and export markets. This increases the diversity of food items and nutritional value available in local markets, as well as the incomes and food-purchasing power of producers and traders involved (see subchapter C.3). It is estimated that at least 600 to 700 non-wood forest products are commercially important and internationally traded, including 500 to 600 medicinal plants. These products have great potential in developing countries where labor is cheap and natural supplies are still ample (Iqbal 1993) (FAO 1988b). Policy measures and cooperation within regional forums can effectively reduce these barriers (see subchapter B.1). For example, in November 1996, Ministers of the Asia-Pacific Economic Cooperation (APEC) forum adopted the Manila Action Plan for APEC. Japan's action plan reduced tariffs to an average of 1.5% on some 6,700 industrial products, including: 1% tariff on forestry products; 4.1% on fishery imports; and 9.3% on agricultural commodities. The Indonesian plan reduces tariffs on almost all commodities to between zero and 5% by the year 2003, although tariffs on agricultural products may be higher than 5% (Widinugraheni and Suryodiningrat 1996).

Developing countries face many barriers and forms of protectionism in the trade of forest products. Trade barriers generally impose a greater burden on smaller enterprises, which have limited capacity to respond and adapt to changing conditions in the international market

The three-year-old World Trade Organization (WTO) completed its inaugural ministerial conference in Singapore in early December 1996. The 8-page conference declaration provided a blueprint for WTO's future agenda, and reaffirmed the commitment of the 128 members to dismantling tariffs and nontariff barriers to achieve the goal of sustainable growth and development. A special plan of action was agreed upon to help the least-developed members respond to opportunities offered by the free trading system. Developed and developing countries still hold polarized positions concerning labor standards (perhaps the most contentious issue), competition and investment policies, and government procurement.

Four major challenges confront WTO:

• continue liberalizing trade in goods and services;
• agree on terms of entry into WTO for China, world's 11th largest exporter;
• deal with contentious issues of foreign investment, competition policy and labor standards; and
• reconcile regional trading agreements within the global free trade framework (The Economist 1996).

Formal WTO trade negotiations on agricultural products are slated to begin in 1999. Negotiations on agricultural trade practices should encourage and support sustainable agriculture. Top priority should be given to food security and the health and nutrition of all people. Environmentally and socially destructive trade practices must be eliminated through open, balanced. non-discriminatory, multilateral negotiations.

6. Information

Several types of information constraints and gaps exist among the linkages between forestry and food security, including:

• Lack of information and understanding among foresters, agronomists, nutritionists, policy makers and planners on the important linkages between forestry/agroforestry and food security.
• Inaccessibility of technical information on these subjects among smallholder farm households and grassroots organizations.
• Lack of market and price information on food products from agroforestry and forestry systems.
• Lack of mechanisms between countries in Asia-Pacific to exchange information and experience on successful approaches to enhance the contributions of forestry and agroforestry to food security.

At the international level, APEC leaders agreed to complete major tariff cuts on information technology by the year 2000, and to call for the conclusion of an information technology agreement by the first World Trade Organization ministerial conference held December 1996 in Singapore (Thaitawat 1996). The conference agreed with this proposal, which may be beneficial in the future for alleviating some of the information constraints.

There are also many contentious issues related to information in the context of WTO and international trade; chief among these issues are intellectual property rights (IPRs). Critics of GATT, predecessor of WTO, charge that IPRs involving ownership and monopoly control over life forms and processes have profound implications on agriculture and food security in developing countries such as the Philippines.

By transforming biological resources from common heritage into patentable commodities, IPRs can erode the economic, ethical and cultural values of farmers in developing countries. The traditional relationship of these farmers with corporations exists on three levels-as supplier of germplasm, as competitor of innovation and rights to genetic resources, and as consumer of products. There is a fear that with patent protection, the farmer will be transformed into a free supplier of raw materials, displaced as a competitor, and become dependent on vital production inputs like seed. This has led to the view by many developing countries and NGOs that "equal provisions under unequal situations will only perpetuate inequality" (Zamora 1994).

7. Industrialization/urbanization

Half of the world's 60 largest cities are in the Asia-Pacific region. The rising tide of rural flight to urban areas is a serious problem in many countries in the region-particularly in the LIFDCs such as China, India, Indonesia, the Philippines and Pakistan. For example, in Indonesia, the total urban population is increasing by about 3 million people every year. If this trend continues, the majority of Indonesians, traditionally an agrarian society, will be living in cities by 2010. Industrial development, in urban as well as rural sectors, is also expanding rapidly in many countries, fueled by domestic and foreign investment. In China, many ancient villages throughout the countryside are dying out as people head for urban areas where industries provide job opportunities. The result: China's big cities are now host to tens of millions of squatters and homeless, most of whom come from impoverished, interior provinces (IHT 1996).

These trends threaten the very foundation of food security-a vibrant agricultural workforce-as able-bodied people leave rural areas and their traditional farming livelihoods. As industrialization and urbanization advance into rural areas, forests, agroforests and farms are converted to factories, residences, offices, shopping complexes and recreational areas such as golf courses. Nowhere is this process more dramatic than in the coastal parts of Guangdong province, southern China, where the traditional, integrated agroforestry systems are being rapidly paved over. Instead of growing their own rice and food, more people in Guangdong will be purchasing their food with increasing incomes.

One possible way to stem rural flight is to develop agribusiness and other rural enterprises with good economic alternatives and incentives for people to remain in rural areas. In Indonesia's second long-term (25-year) economic development plan, rural agribusiness development is one of the most prominent features and will receive substantial public and private investment. In Thailand, fruit orchards and orchid gardens are now employing tissue culture and other "high-tech" means to maintain their comparative advantage, and contribute to food production, employment and income generation in the countryside.

8. Food deficits/malnutrition

The chronically undernourished in the Asia-Pacific region number about half a billion, accounting for two-thirds of all the world's malnourished people. There is a great need to implement the national plans of action on nutrition in accordance with the commitments made at the 1992 International Conference on Nutrition (FAO 1996).

Forest and agroforestry products play an extremely important role in helping to ensure adequate nutrition. They provide food in many forms, contribute to dietary diversity, supply nutrients and micro-elements, and minimize seasonal or emergency food shortages. However, foresters and nutritionists seldom work together to integrate these aspects into field projects. Ogden (1990) developed a methodology to guide forestry project planners in the following steps:

• Obtain essential background information on nutrition-related factors such as vulnerable groups, use of trees and forests, food supply, income, the environment and gender roles.
• Identify nutrition objectives that are compatible with the overall objectives of a given forestry project, the time-scale, and, most importantly, the community's needs and aspirations.
• Select appropriate indicators to monitor and evaluate the nutritional impact of the project. This requires creative thinking as traditional nutritional indicators are not appropriate, and there is a need to select indicators based on how the community defines the criteria for success or failure of the project.

In the Pacific, staple tree crops-such as coconut, breadfruit, bananas and plantains-and the wide range of fruit and nut trees are critical to the nutritional wellbeing of island peoples. The replacement of these foods by processed, imported products has led to some of the highest rates of malnutrition and nutrition-related, non-communicable diseases in the world (Thaman 1996).

Food security is intimately linked with population growth. In Indonesia, the current population growth rate is about 1.6% per annum. The food security index indicates the capability of Indonesia to produce rice to meet domestic demand. Food availability is considered secure if the index is higher than 1.00. In 1988, the index was computed to be 1.05; this has declined to 0.98 in 1995, when more than 2 million tons of rice were imported to meet domestic needs. During the same period, the labor force participation rate in Indonesia-defined as the ratio between the labor force (those 15 to 64 years of age) and the total number of people over 10 years old-gradually declined by about 0.2% per year Adnyana 1996).

B. Policy status

1. Regional and international cooperation

In Asia-Pacific, international, regional and sub-regional forums are important mechanisms for promoting cooperation among nations. Four sub-regional forums play key roles:

• Association of Southeast Asian Nations (ASEAN) currently comprises seven members: Brunei, Indonesia, Malaysia, the Philippines, Singapore, Thailand and Vietnam. Burma, Cambodia and Lao PDR are to be admitted as full members "simultaneously," possibly in 1997 (Sawatsawang 1996).
• Mekong River Commission (MRC) involves countries in the lower Mekong basin: Cambodia, Lao PDR, Thailand and Vietnam.
• South Asian Association for Regional Cooperation (SAARC) members are: Bangladesh, India, Nepal, Pakistan and Sri Lanka.
• South Pacific Forum (SPF) covers the South Pacific countries, and includes the South Pacific Commission (SPC), which supports technical cooperation activities among the countries.

These forums can be instrumental in shaping the political, economic, security and strategic dynamics of the region. ASEAN is emerging as an independent center of power. Economically, ASEAN states represent the fourth-largest trading entity in the world, after the U.S., Japan and the European Union. A strong Southeast Asian consciousness and cohesiveness is clearly emerging. However, formidable challenges loom ahead as traditional values and ideologies are being eroded by the twin forces of modernization and globalization (Dupont 1996). By establishing mechanisms such as an ASEAN Free Trade Area and visa-free entry for ASEAN nationals, many trade and movement barriers are lowered. These mechanisms will be increasingly important as ASEAN members utilize long-term food trade arrangements and "comparative advantage" to meet national as well as regional food security goals. Similar means can be used to promote intercountry trade (including barter) of food products from forestry and agroforestry.

Two international forums are gaining increasing influence within the Asia-Pacific region:

• Asia-Pacific Economic Cooperation (APEC) forum comprises: Australia, Brunei, Canada, Chile, China, Hong Kong, Indonesia, Japan, Malaysia, Mexico, New Zealand, Papua New Guinea, the Philippines, Singapore, South Korea, Taiwan, Thailand, and the United States.
• World Trade Organization (WTO), headquartered in Geneva, is charged with implementing an open multilateral trade system based on the commitments made at the Uruguay Round of Multilateral Trade Negotiations. There are 128 members and some 30 other countries who wish to join, including China and Russia. WTO also has responsibilities for issues of trade and the environment; such are chiefly related to agricultural production. Future environmental activities may be relevant to certification of forest products.

APEC strives for cooperation in economic and technical development. Such can only be achieved if citizens within the region were to benefit from and endorse the APEC framework of principles for economic cooperation and development. APEC Ministers were instructed to "give a human face" to development and accord the highest priority to six themes: develop human capital; strengthen economic infrastructure; develop safe, stable and efficient capital markets; safeguard quality of life through environmentally sound growth; and strengthen the dynamism of small and medium enterprises (Suryodiningrat and Widinugraheni 1996). Private sector investment and cooperation is being encouraged.

The first WTO ministerial conference was held in Singapore in December 1996. APEC members successfully urged the conclusion of an information technology agreement at the WTO meeting, which will substantially eliminate tariffs by 2000, but with flexibility for product coverage and a looser time frame. They also encouraged progress on negotiations on protocol issues and market access, with a view to achieving universality of WTO membership. Two APEC members-China and Taiwan-are not yet included in WTO.

2. Economies in transition

Four of the LIFDCs in Asia-Cambodia, China, Lao PDR and Mongolia-are in the process of transforming their centrally planned economic systems into market-oriented economies. The aim is to utilize market forces to stimulate economic growth and more efficient production and distribution of goods and services. Macroeconomic policy reforms also enhance the capacity to earn foreign exchange through increased competitiveness in export markets. A prime example of this is the increasing export of bamboo shoots and other bamboo products from China.

In countries with economies in transition, the forestry sector is also strongly affected by economic reforms, which pose great challenges and opportunities. There are important implications for forest policies and institutions as well as economic aspects of market orientation in forestry, including the following key issues that emerged from an international workshop on this theme, held in Fuzhou, China in 1994 (McNicoll and Durst 1995):

Forestry policies and institutions under a market-oriented economies:

• Modification of adaptation of forestry policies and legislation to support market orientation and reform
• Consideration of land and tree tenure in the reform process; and customary rights under market-oriented tenure structures
• Impact of market reforms on local communities
• Appropriate roles of government forestry agencies
• Reform of state-owned forest enterprises
• Sectoral policies and planning and investment development needs require assistance from fao or other cooperation agencies

Economic aspects of market orientation in forestry:

• Use of incentives to promote improved forest management under a market economy
• Criteria for attracting and assessing forestry investments
• Extent of value-added processing to be developed
• Enhancing competitiveness under a market economy
• Pricing and valuation of forests and forest products
• Forestry market and price information needs

3. Policy disincentives and incentives

Forestry and agroforestry development in support of food security is greatly affected by policies, rules, regulations and incentives (or disincentives) enforced by the state and community. Formal legislation may exist at the local, national or international level and often act as disincentives to forestry investment, production and marketing. For example, national forestry laws may restrict the harvest of certain "royal" tree species; declare certain species as exclusive property of the state (e.g., sandalwood in India); or require mandatory permits for the transport of all wood regardless of where the trees were grown or by whom. Teak in most countries requires a permit to be felled and transported, even if grown on private land. This is a constraint for farmers who are often exploited by unscrupulous officials who issue the "free" permits. However, these laws are necessary to protect against unauthorized felling in public forests.

Forest policies are increasingly described as "antiquated" or as "colonial legacy," and were conceived in an era when forest management objectives and conditions were radically different from the present. However, even within the existing, outdated policy framework, there is scope for interpreting and implementing elements that can act as incentives to local communities and farmers.

The key is the willingness of governments and forest departments to create an enabling environment, within the limits set by prevailing policy and law. Understanding these conditions helps identify barriers and opportunities, incentives and disincentives for forestry and agroforestry (Lai and French 1994).

Apart from formal legislation and policies, traditional customs and practices govern management of agricultural and forest lands. User rights are particularly important for farmers who live on the forest fringes, and are dependent on forests for food, fuelwood and other products. Traditional rules and rights are governed by social norms and customs, and can be very effective incentives for improving production as well as conservation of natural resources (French 1995).

C. Major watersheds and forested areas

1. Sustainable water

In the 21st century, maintaining a sustainable and adequate supply of water for household and food production needs in the Asia-Pacific may become the most crucial challenge facing farmers, urban dwellers and governments. This scenario threatens the fundamental basis of food production and food security.

The role of protective forest cover in critical watersheds to ensure sustainable supply of clean water is generally well-understood. However, the forces driving deforestation and agrodeforestation (see subchapter A.2) destroy or degrade many watershed forests and tree resources. This has a dangerous and adverse effect on the water resources that are vital for agriculture, and, indeed, life in general. Without forest or tree cover, soils lose much of their ability to absorb and retain water, leading to increased runoff, soil erosion, siltation and flash flooding. Today, available water supplies for each Asian are less than half of the world's average (Mercado 1996).

The problem of sustainable water is obviously more acute in dry zones or agroecosystems with prolonged dry seasons during the year. With growing population and urbanization, water shortages are also severe in and around many cities and towns in the region. There is increasing incidence of: "water-mining" of aquifers; resultant salinization of groundwater; lowering of water tables in areas such as western and southern India, northern China and central Philippines; and water use conflicts among agricultural, industrial, residential and recreational (e.g., golf courses) interests. Smallholder farmers often lose out to more powerful competing interests, including large-scale agricultural enterprises.

Cebu Island in central Philippines, with the highest population density in the Philippines, is a prime example of the evolving water crisis confronting many areas in Asia-Pacific. The lush molave forests that Ferdinand Magellan encountered when his ships dropped anchor in Cebu in 1521 have been cut. Today, the landscape is characterized by eroded hillsides, silted rivers, denuded watersheds and parched farms. Cebu's vibrant economic boom has further contributed to escalating water shortages. By 1990, over 150,000 cubic meters per day were pumped out from their limited aquifers. It is estimated that seawater intrusion already exists in a third of Cebu's aquifers (Mercado 1996).

Initiatives such as CUSW (see box) are important for involving civil society, as well as a wide range of interests and users from the public and private sectors, to address the complex problems related to watershed management. Hopefully, CUSW can serve as a model of an effective, broad-based forum to ensuring sustainable water for future generations.

2. Participatory watershed management

The CUSW initiative cited above is a good example of participatory watershed management, albeit in a more urban setting. In recent years, participatory approaches to watershed development have grown in response to the technical, social and economic failures of many past efforts in soil and water conservation. In the past century, most policies and projects had taken the view that farmers generally mismanage soil and water, and deforest and degrade watersheds. However, soil and water conservation practices based on imposed technological interventions have usually not delivered the expected environmental or economic benefits. Designing and implementing interventions without involving local people can only succeed with either coercion, or through the use of financial, food or legal incentives. Approaches and subsidies of this nature are not sustainable, and inappropriate structures and technologies frequently resulted in more erosion (Pretty and Shah 1994).

Participatory watershed development differs from conventional approaches in several important ways, including the following:

• Local communities are fully involved in problem analysis and decision making.
• External support organizations act as facilitators and catalysts to help local institutions.
• Farmer-to-farmer extension is recognized and used as a key process.
• Project staff do not take a blueprint approach, but employ flexibility based on the individual needs of farmers.
• Emphasis is on sustainability and equity of improvements rather than on short-term benefits; on self-help mechanisms rather than payments or subsidies (Hinchcliff et al. 1995).

There are a growing number of small-scale projects that are sufficiently successful to warrant application of their experiences on a wider scale. Successful joint watershed management efforts generally share common elements. First, all emphasized the use of locally-adapted, resource-conserving technologies that provide immediate returns to farmers. Second, they encouraged and supported actions of local groups or communities, rather than working with individual farmers. Third, they involved collaboration among farmers, supportive government units and/or NGOs, rather than working in isolation (Hinchcliff et al. 1995).

A recent case study on successful participatory watershed management in protected areas in northern Thailand identified three key factors: 1) secure land tenure or usufruct rights; 2) strong farmers' organizations as elements of intervillage or watershed network, and human resource development at village and watershed level; and 3) sufficient capital investment to support resource management by the local people (Emphandhu et al. 1996).

On Java Island, Indonesia, as in other high-density areas of the tropics, watershed management projects often regard downstream protection as a primary objective. In practice, the on-site cost of erosion due to reduced productivity is often higher than the downstream damage through siltation. Magrath and Arens (1989) estimated that soil erosion on Java costs the economy between US$340-405 million annually, with US$315 million on-farm losses and US$25-80 downstream damage. Most soil conservation projects on Java focus on construction of new terraces and terrace rehabilitation, with relatively high subsidies paid to participating farmers. Reij and Crichley (1994) suggested that the strategic emphasis should be on improving existing terraces, rather than constructing new ones. They also recommended, as possibly the single most important and simplest technical intervention, the improvement of planting of fodder grasses on terrace lips or risers in conjunction with stall-feeding of livestock.

3. Non-wood forest and agroforestry products

Forestry and agroforestry are linked to food security in three major ways:

• the environmental links between forestry/agroforestry and food production, and the role of watershed forests in conserving soil and water resources;
• the household nutrition links in terms of consumption of forestry and agroforestry products; and
• the socioeconomic links in terms of generating income and employment to support household food security.

In recent years, there has been increasing recognition that non-wood forest products can: (a) provide important community needs for improved rural livelihood; (b) contribute to household food security and nutrition; (c) help to generate additional employment and income; (d) offer opportunities for processing and value-added enterprises; (e) contribute to foreign exchange earnings; and (f) support biodiversity conservation and other environmental objectives (FAO 1995b).

Non-wood products derived from forestry and agroforestry have direct impacts on food production and security. The production of fruits, nuts, a wide variety of other foodstuffs, medicinal products, wildlife and fishery products, among others, directly provide important contributions to the household diet, nutritional status and economy. Non-wood forest and agroforestry products also provide opportunities to develop a diverse range of income and employment generating activities. While precise data are lacking on this aspect, it is clear that in LIFDCs in Asia-Pacific, millions of people depend on forestry and agroforestry for income and employment, and particularly from the collection, processing and marketing of non-wood products. Income earned from these activities enables the family, usually the woman who is responsible for food supply, to purchase food and invest in future production.

A major strategy for future action is to focus on enhancing local livelihood through local enterprises based on forest and agroforestry resources. Capturing an appropriate share of a product's added value closer to the source helps ensure the equitable distribution of benefits through the marketing chain as well as the proper valuation of the resource in management decisions (FAO 1995c).

In 1994, over 70 participants from more than 20 countries in the region met to consult on the social, economic and cultural dimensions of non-wood forest products. There was a solid consensus on the need for local community control and management of non-wood forest resources. The consultation highlighted several successful or promising non-wood forest product initiatives. These included efforts to improve processing and/or marketing of: forest nuts in the Solomon Islands; preserves and jams from forest fruits in the Philippines; ecotourism in Fiji; medicinal plants in India, Nepal, Sri Lanka and Indonesia; rattan in the Philippines; non-wood mangrove forest products in Bangladesh; and bamboo products in China.

The consultation also recognized the risks associated with the accelerated development of non-wood forest products. The major risks are: over-exploitation of resources; competitive exclusion of the poor as demand and prices increase; boom-and-bust product cycles; over-dependence on a single commodity or product; and threats to traditional cultures from new orientations toward cash economies (Durst and Bishop 1995).

D. Dry zones

1. Land and production limitations

Arid and semi-arid lands cover about one-third of the land surface of the Earth, and are home to almost 1 billion people. With increasing population growth and food demand, food production is expanding into fragile dry regions where rainfall is less; soils are more sandy, less fertile or shallower; or slopes are steeper.

The principal agents of dryland deterioration are water and wind erosion resulting from tree and shrub cutting for fodder and fuelwood, heavy grazing of rangelands, and cultivation of sandy or sloping lands. Salinization caused by improper land management is the major problem in irrigated lands (Dregne 1989).

In Asia, drylands used for rangeland and agricultural purposes are estimated at almost 1,900 million ha, 70% of which are in degraded condition (UNEP 1991). Among the LIFDCs in Asia-Pacific, dry zones are a dominant feature in major portions of China, Mongolia, India, Pakistan and Sri Lanka. Some arid and semi-arid lands are also found in other countries, for example, the Nusa Tenggara island chain in eastern Indonesia. The low suitability and productivity of drylands for food production makes agricultural expansion in these areas very problematic and environmentally risky. However, given the very low agricultural land per capita in these LIFDCs (Table 2), coupled with rising populations, there will be continuing pressures to expand food production in dry areas.

Table 2. Trends in development of rainfed and irrigated lands in LIFDCs.

Source: FAO 1995d; data not available for Kiribati and Tuvalu

2. Rainfed areas

The arid regions of India cover more than 30 million ha, about 12% of the country's total land area (Harsh et al. 1993). Hostile environmental conditions prevail, but more than 40% of these arid and semi-arid lands are under cultivation. Growing only annual crops on such lands has been found to be uneconomical and extremely risky. Therefore, most farmers raise livestock as a subsidiary occupation, and grow shrubs and trees in the cultivated tract to cover the risk of crop failure as well as to provide fuel, fodder, fruit, timber and other products. These age-old agroforestry systems are based on hardy tree species such as Acacia spp., Prosopis spp. and Ziziphus spp. (Shankarnarayan et al. 1989).

Over half of the lands in India, more than 150 million ha, are under rainfed farming systems of varying productivity. These rainfed lands are under growing human and livestock population pressures. This has led to agricultural expansion into marginal lands, such as drylands, overcutting of trees and shrubs, and overgrazing of pastures. To respond to the problems facing rainfed areas, the Government of India launched the National Watershed Development Programme for Rainfed Areas (NWDPRA) in 1991, with an allocation of some US$600 million. The programme employs an integrated watershed development approach to promote mixed farming systems under complex, diverse and risk-prone environments in order to restore ecological balance and improve the socioeconomic welfare of the watershed communities. About 2,500 pilot, micro-watersheds (ranging in size from 500 to 5,000 ha) have been selected in development blocks where less than 30% of the area has irrigation facilities. These watersheds will be surveyed, planned and developed with the active participation of local communities and NGOs (Seth 1992).

A rapid review of NWDPRA in 1995 examined the program's process and approach, and offered recommendations on: (a) integrating promising indigenous technologies; (b) improving support for effective farmers' organizations; (c) developing methods for participatory monitoring and evaluation; (d) providing secure tenure for managing common property resources; and (e) establishing national working groups to strengthen NWDPRA (Sharma and Stainburn 1995).

Rapid agricultural expansion and deforestation in China have contributed to increased soil erosion, runoff, siltation of rivers and flooding. Reservoirs along the Yellow River basin (draining the heavily eroded Loess plateau) and the Yangtze River are showing alarming siltation rates that affect their storage capacity. In response, the Government of China has launched massive forestry and agroforestry programs, including the "Three North" Shelterbelt Development Programme, located in the arid/semi-arid zone of northwest, north central and northeast China. Started in 1978, it is now perhaps the world's largest agroforestry initiative. Large-scale farm shelterbelt systems have been established, sand-dunes have been fixed, forests have been planted for soil and water conservation, and other watershed rehabilitation works have been carried out, with generally positive results. An estimated 11 million ha of agricultural land have been protected, 9 million ha of desert reclaimed with vegetation, and over 7 million ha of eroded plateau area stabilized (reducing by 10% the erosion flow to the Yellow River).

The improved conditions have contributed to overall grain production increases of 10-30% and grass production improvements of over 20%. Yield increases due to shelterbelt establishment averaging 16% for maize, 36% for soybean, 43% for sorghum and 44% for millet have been reported (Ministry of Forestry, PRC 1992, Liu and Pang 1991, Ball et al. 1995).

3. Irrigated production systems

Land under irrigation comprises roughly one-third of the total agricultural land area in the Asia-Pacific region. Irrigated areas in Pakistan, China and India make up 80%, 52% and 28%, respectively, of the total cultivated land (Table 2). Many of these irrigated production systems occur in arid and semi-arid regions. Salinization of irrigated lands is a continual problem. All irrigation water contains some amount of salt. If soil drainage is impaired, soluble salts will accumulate and can harm crops. Poor drainage is the principal cause of the extensive areas of salt-affected soils in Pakistan and China. Waterlogging, siltation of canals and diseases are other problems associated with irrigated areas (Dregne 1989). Inefficient irrigation systems can also result in tremendous waste of scarce water resources, losing up to 45% of the water that flows through irrigation canals (Mercado 1995).

Despite these problems, irrigated production systems are vital to food security in dry zones. Over the past decade, the irrigated cropping area in Pakistan, China and India has increased faster than the total area under cultivation. Both the production and productivity of irrigated crops have increased at a faster rate than rainfed crops. For example, between 1950 and 1990 in India, the yield of irrigated wheat and sugarcane increased by 200-300%, whereas rainfed pulses and oilseeds only registered 15-50% improvements in productivity during the same period. One potentially risky trend is the move toward irrigated monoculture crops such as: intensive paddy and wheat rotation in northwest India; irrigated teak and eucalyptus plantations; and fruit orchards under sprinkler irrigation. While these irrigated monocultures have certain economic advantages, they are also ecologically susceptible to pests and diseases (Seth 1992).

The world's largest canal irrigation system, over 62,000 km in length, is found in the flat Indus Plain in central Pakistan. This system makes cultivation possible despite the scanty and erratic rainfall and extreme temperatures. Agroforestry is becoming increasingly important in Pakistan. There are excellent prospects to extend agroforestry to uncultivated sloping lands and saline/waterlogged fields. Agroforestry systems now cover only about 2% of the cultivated area in Pakistan, but produce 9 times more fuelwood and 3 times more timber than state forests. Irrigated agroforestry systems also contribute to food security. Fruit-tree planting is an age-old tradition on irrigated farms in mountainous areas. The date palm is the most frequently found tree species on irrigated farms in the northwest Frontier Province. Non-wood forest products are also important sources of food and cash, providing up to one-third of the income of those involved in collection and processing (Koppelman et al. 1996).

4. Role of forestry and agroforestry

From the foregoing discussion, the importance of forestry and agroforestry to environmental amelioration and food security in dry regions cannot be overstated. The key roles may be summarized as follows:

• Forests and agroforests in dry regions directly provide supplemental food and nutrition to households, including emergency foods during periods of drought.
• Appropriate agroforestry tree species can improve the micro-environment for crop cultivation in hot, arid zones by reducing ambient and soil temperatures, mitigating soil erosion, and improving soil fertility.
• Trees planted or protected by farmers help offset the risk of crop failure in dry and marginal lands, and provide cash, food, fuel, fodder and wood for household needs.
• Non-wood forestry and agroforestry products provide a wide range of opportunities for generating income and employment related to the collection, processing and marketing of these products.

There is a need to integrate appropriate forestry and agroforestry production systems into the overall land-use or watershed management approaches being implemented in dry zones and rainfed areas of LIFDCs.

E. Temperate zones

Among the LIFDCs in Asia-Pacific, significant temperate areas are found in Mongolia, the plains and deserts of northern China, and the mountainous Hindu Kush-Himalayan region extending across China, Bhutan, Nepal, India and Pakistan.

1. Soil erosion and fertility decline

Many rich agricultural regions are found in these temperate zones. However, soil erosion and declining fertility are two of the most pressing problems facing farmers in these areas. For example, the 10 large plains of northern China-representing 15% of the total territory, 45% of the arable land and half of the population-is the most important base of grain, cotton and oilseed crops in the country. In the past, the sparse vegetation on the plains provided very limited protection from the frequent sand storms, droughts, floods and other natural disasters. Wind and water erosion of soil was heavy, resulting in low and unstable grain production and shortage of fuelwood and timber (Ministry of Forestry, PRC 1992)

In most parts of the Hindu-Kush Himalayan region, mountain agriculture is showing signs of decline due to the rapid erosion of the natural resource base that supports agriculture. In the past, mountain agriculture was sustained through the organic integration of crop production, horticulture, forestry and animal husbandry. However, increasing population, market forces and public interventions have caused organic disintegration, leading to reduced area under forests and trees, reduced agricultural diversity, and increased runoff and soil erosion. In spite of their ability to raise crop productivity under favorable circumstances, the Green Revolution technologies and high-yield varieties offer no solutions to the basic problem of natural resource degradation (Denholm 1991).

2. Agroforestry to sustain food production

Appropriate agroforestry systems and practices may help arrest soil erosion and improve soil fertility in temperate zones. China has made great strides in afforestation, and the establishment of windbreaks and shelterbelt systems in the plains. Since 1978, shelterbelts have been planted on 28 million ha of farmland (about 75% of the total area suitable for shelterbelt establishment) as well as on 0.4 million ha of pasture land (almost 10% of the total). After 40 years of work in Henan province, 4 million ha of farmland are now under shelterbelt protection. A recent survey indicates that shelterbelts can reduce wind speed by 30-40%, lower evaporation rates by 20-25%, increase relative humidity by 5-15%, and lead to 10-20% increases in agricultural crop production (Ministry of Forestry, PRC 1992).

In Hebei, Shandong, Henan, Anhui, Shanxi and Jiangsu provinces, Paulownia-based agroforestry systems (e.g., intercropped with wheat, maize, cotton) have been established by farmers on well over 3 million ha, or almost 60% of the total arable land in the six provinces (Zhong 1991). Farmers prefer moderate spacing of trees as it provides higher economic returns than monocropping, and, at the same time, produces enough food to meet their requirements. Economic benefits from Paulownia intercropping may be 15-30% higher than monocultural systems (Zhu et al. n.d.). With the market economy orientation, there are excellent prospects for the export of Paulownia wood and processed products to Japan and other countries.

There are over 3 million ha of poplar (Populus spp.) plantations in China; most occur in intercropping systems in the plains. Poplar-based agroforestry systems were reported to be beneficial for crop yields, particularly bean, peanut, wheat and rapeseed (Chen 1994).

In the Hindu Kush-Himalayan region, based on the potential of resource-centered technologies to address the twin goals of productivity and conservation, agroforestry is vital to the sustainable development of mountain agriculture. The most commonly encountered forms of mountain agroforestry are agrisilvopastoral systems that combine trees and crops with animals or pasture. These include the practices of: homestead agroforestry, livestock grazing on fallow croplands; establishment of fodder trees around fields; and use of woody shrubs and hedges for mulch, browse and green manure. Two key factors affect the evolution of mountain agroforestry:

• the crucial role of biomass derived from perennials, both in farming systems and the subsistence strategies of hill farmers; and
• the changing patterns of availability and access to tree products with increasing pressure on common property lands.

An essential component of traditional mountain farming systems is maintaining soil fertility through internal resource regeneration, namely the addition of farmyard manure and forest litter. The most significant and widespread strategy for soil fertility is the transfer of nutrients from forests and grazing lands via the form of livestock manure. Therefore, the fodder element is the linchpin in the farming systems of the hills and mountains. Fodder from trees and grasses constitutes well over half of the diet of livestock on an annual basis. The indigenous agroforestry systems, programmes, policies, institutions and problems and constraints found in mountainous areas of Bhutan, China, India, Nepal and Pakistan are highlighted in an ICIMOD publication (Denholm 1991).

F. Small islands

Most nations in the Pacific comprise relatively small islands and atolls. Among the currently classified LIFDCs in the South Pacific, Kiribati, Samoa, Solomon Islands, Tuvalu and Vanuatu are characterized by the presence of small islands. Among the LIFDCs in Southeast Asia, Indonesia and the Philippines are vast archipelagoes composed of over 13,000 and 7,000 islands, respectively, many of which are small islands, including those not inhabited. Among the South Asian LIFDCs, the Maldives encompasses small atolls. Small islands represent a wide range of agroecological and socioeconomic conditions. Due to their small size and, often, large distance from other land masses, constraints that affect food security on small islands can be difficult to address.

1. Traditional agroforestry systems

Traditional agricultural production systems in the Pacific Islands-as well as in most other small island environments-are forms of agroforestry, where trees are planted and/or protected for the great variety of their functions and products, including food. All Pacific countries have highly developed agricultural practices. For example, agroforestry in Micronesia (in Yap) and in Tonga have been developed to high degrees of sophistication. In Pohnpei Island, Federated States of Micronesia, indigenous agroforestry (pahnwel)-characterized as semi-permanent, multistory and tree-dominated-is the major agricultural land use, and covered one-third of the island as of 1984 (Raynor 1991).

In traditional Pacific island societies, forestry, agriculture, housing, health and nutrition, education, trade and industry were not compartmentalized into economic sectors. Instead, they were managed as components of integrated production systems adapted to the environmental, economic and cultural needs of local communities (Thaman and Whistler 1996).

Food or fruit species are the most common trees and shrubs in villages and houseyard gardens, in permanent village tree groves, and as protected or planted intercrops in food gardens. They include a wide range of: coconut palms, banana and plantain cultivars, breadfruit, edible pandanus varieties (especially on atolls), fruit trees, nut and seed trees, and kava (root used for traditional alkaloid social beverage). Most of these species are either aboriginal introductions (pre-European contact) or, in some cases, indigenous. Many recently introduced banana varieties are now commonly grown. In most coastal areas and on atolls, the coconut is ubiquitous within gardens and in areas under short-term fallow (Thaman and Whistler 1996).

Foods from trees are of immense value as staples, supplementary foods, snacks, or famine food. As discussed in subchapter A.8, staple foods from tree crops are critical to the nutritional wellbeing of Pacific island peoples. Replacement of these foods by highly-processed and imported foods has resulted in widespread malnutrition in the region.

2. Atolls

Among the LIFDCs in the Asia-Pacific region, the Maldives, Kiribati and Tuvalu are made up of small atolls and raised limestone islands scattered over great expanses of open ocean. An atoll is a circular reef of coral islets, partially covered by seawater enclosing a lagoon. The islets are usually less than 3 m in height above sea level. The land surface area-as in the Western Isles of Kiribati-may more than double when the tide goes out. Skill in managing the fragile freshwater, which forms fragile lenses sitting on top of the salt groundwater, literally determines life and death (Richardson and Richardson 1996).

The atoll environment is very harsh. Derived from coral limestone, atoll soils are shallow, alkaline, coarse-textured, and lacking in most nutrients required for plant growth. Soil water-holding capacity is very low, and plant nutrient is highly dependent upon the humus cycle and maintaining the vegetative cover. The islands of Kiribati and Tuvalu are rarely more than 3 m above mean sea level. Both countries are extremely vulnerable to coastal erosion, salt-water intrusion, high spring tides, storm surges and seismic sea waves. Therefore, coastal forest protection and reforestation are high priorities. Soil resources are among the most infertile in the world, and there are virtually no surface freshwater resources (Thaman and Whistler 1996).

Despite inadequate land, soil and water resources as well as relatively high population pressures, I-Kiribati (people of Kiribati) and Tuvaluans have developed sophisticated subsistence agroforestry systems. These systems are based on coconut, breadfruit, pandanus, native fig, bananas (on the wetter islands) and giant swamp taro, cultivated in excavated pits dug to the freshwater lens. This type of pit cultivation, utilizing leaves of salt-tolerant coastal trees and plants as mulch and fertilizer, is also applied to important staple tree crops to ensure their survival in the atoll environment. Thaman and Clarke (1987) listed 80 ecological and cultural functions and uses by atoll societies. The introduction of a cash-based economy has affected the lifestyle of atoll dwellers and lessened their dependence on trees. Growing population, urbanization and market-oriented economy have led to a declining area available for growing trees and food crops (Teunissen et al. 1996).

3. Potential to enhance food security

Forestry and agroforestry development will continue to be vital for the food security of small islands that constitute South Pacific nations. Despite changing demographic, social and economic conditions, the foundations and potential for agroforestry to sustain food production remain strong.

Working together with farmers, participatory farming systems research and extension can help find answers to some pressing constraints to increasing and sustaining food production. For example, in most Pacific countries, tubers are the staple food: taro is eaten extensively, and yams are reserved for special occasions. Farmers traditionally remove all organic material from their taro gardens because they believe such material will attract the Papuana taro beetle, a major pest. On-farm trials are testing the impact of different trees such as Gliricidia sepium, Erythrina subumbrans and Hibiscus tiliaceus on the yield of taro, on weeding and labor demand, and on the use of mulch to reduce taro beetle damage. Previous studies have shown that agroforestry can be labor-saving if managed properly, with tree coppicing replacing weeding as the major (and much easier) maintenance task. Light shading regimes from intercropped trees can actually increase taro yields and make weeding work less onerous by providing a cooler microclimate (Lai 1996a).

There is also great potential to link nutritional aspects with agroforestry and horticultural efforts. For example, a project supported by UNICEF in nine Pacific countries helped raise awareness about nutritional problems such as underweight children, and vitamin A deficiency and non-communicable diseases in adults. This was combined with project efforts to integrate vegetables into family gardens to improve household nutrition (Barker 1996). Surplus production can also be marketed to earn cash to buy food and other essential items. In Kiribati, a decade of work supported by the Foundation for the Peoples of the South Pacific has led to the improvement of the nutritional status of women and children through the local production of vitamin A-rich vegetables (Thaman and Whistler 1996).

The evolution of agroforestry in the Pacific Islands may be considered over five distinct periods:

• agriculturalisation of the forest
• indigenous agroforestry enrichment and deforestation
• colonial agroforestry enrichment and agrodeforestation
• post-World War II agroforestry enrichment and accelerated agrodeforestation
• 21st century "evolutionary agroforestry"

"Evolutionary agroforestry" may be viewed as the gradual development, as a result of growth and change, of highly-developed land-management systems in which the deliberate incorporation and protection of trees have been combined with the production of plants and/or animals to ensure their long-term productivity and sustainability (Thaman 1996). This process stresses the need to protect or enhance existing, traditional agroforestry systems and indigenous knowledge in the Pacific Islands.

G. Urban agroforestry

1. Importance to household food security

By the year 2020, Habitat projects that two-thirds of the world population will live in urban areas-towns and cities. It is conceivable that towns and cities in many countries could produce up to half the food they need by that time. Urban agriculture produces 3-15 times more food than common rural methods, and is often more organic and sustainable. It utilizes urban waste, which is 70% organic, and the urban farmers' more intensive methods use less land and water per unit of production than industrial agriculture (Smit 1996a).

A recent UNDP study reveals that urban agriculture may produce as much as 15% of the world's food, although it has been often ignored or discouraged in the past. By examining urban farming practices in 100 cities in 30 countries, the study concludes that one in three of the world's urban residents grow food, either for consumption or sale (Smit 1996b).

Urban agroforestry systems and home gardens provide multiple products and services beyond food. Trees in the urban landscape provide shade, aesthetic and environmental benefits, serving as a "green lung" to mitigate against air and noise pollution.

2. Relevance to Asia and the Pacific

In the Asia-Pacific region, urban environments are often overlooked in developing food security strategies. Farmers in Calcutta, one of the world's most crowded cities, produce over a third of the fish and vegetables consumed by city residents. In China, 40% of the jobs in cities are agriculture-related, and many cities, such as Shanghai, are self-sufficient in vegetable production.(Smit 1996a). Beijing is an urban forestry success story. The municipality's green cover of trees and shrubs has increased from 3% of the total area in 1949 to 26 % in 1993, or about 6 m² per Beijing resident (FAO 1993b).

In the Pacific islands, home gardens are a prominent feature of the urban landscape. Random surveys of urban home gardens in capital cities indicate more than 85 different species of food plants are cultivated. A variety of food tress, non-tree staples, supplementary food plants, medicinal plants and non-food species are intercropped in these gardens (Thaman and Whistler 1996).

In addition, many households cultivate "unused" open areas in cities, such as undeveloped road frontages, empty lots, river and canal banks, and construction sites. In Port Moresby, more than one-third of families cultivated idle land in addition to their home gardens. In Suva, Fiji, more than 20% of all households practiced agroforestry in undeveloped urban areas and alongside roads (Thaman 1987).

3. Future prospects

As discussed in subchapter A.7, half of the world's 60 largest cities are located in the Asia-Pacific region. In the early decades of the 21^st century, it is likely that the majority of people in the region will be urban dwellers.

Given its multiple benefits and products, urban agroforestry should be seriously considered and integrated into urban planning efforts throughout Asia-Pacific. It should also be one of the major food security strategies to be adopted and supported by governments as well as the private sector and civil society. In addition to its enormous food production potential-meeting up to half of the food requirement in cities and towns-urban forestry can provide the essential greenness and spiritual refuge that will make places like Bangkok, Jakarta and Manila livable into the next century.