RSS | Register/註冊 | Log in/登入
Site search:
Home>FFTC Document Database>Extension Bulletins>Agro-Ecological Analysis for Agricultural Development in Indonesia
Agro-Ecological Analysis for Agricultural Development in Indonesia
Istiqlal Amien
Center for Soil and Agroclimate Research
Jalan Juanda 98
Bogor 16123, Indonesia, 2002-10-01


Current land allocation in Indonesia is unlikely to support sustainable utilization, neither is it able to adjust to a changing global economy. Using advanced analytical methods, the available information on land resources now can be properly utilized to reevaluate appropriate agricultural land use. An agroecological approach, using a minimum data set covering terrain, soil and climate, can delineate land resources for sustainable utilization such as annual crops, agroforestry, perennial crops and forests. This approach was employed in land resource analysis for Java, Sumatra, Kalimantan, and Sulawesi. The results indicate that in Sumatra, Kalimantan, and parts of Sulawesi, substantial areas of land that are suitable for agriculture are currently utilized for forestry. In Java and parts of Sulawesi, on the other hand, lands that should be forested is cultivated. Using an expert system as a decision support tool, appropriate production systems as well as crop choices for a particular region can be assessed. Agricultural land can be divided into three categories, used respectively for agricultural intensification, expansion, and rehabilitation through diversification programs.


Although Indonesia has had a high rate of growth in industry and services during the last few decades, over half the population is still living in rural areas. The increasing agricultural population and the reduced economic growth rate in recent years has been putting considerable pressure on arable land. Much of the land has become degraded because of misuse and over-exploitation. Unless this trend can be reversed, renewable resources and their sustainable utilization are in jeopardy.

Current land allocation is mainly a continuation of what was established by the colonial Dutch East India Company, and is unlikely to conform with sustainable land use. When Europe was the main destination of Indonesian agricultural exports, it was logical to utilize land in coastal areas close to harbors, unless the crop required the cooler climate of the highlands. As populations increased, farmers were forced into upland areas. With the monsoon rainfall pattern of the tropics, this type of land is disrupting the water cycle, and already causing frequent floods and droughts.

Neither will small-scale subsistence farming be able to adjust to the changing global economy. Trade and transportation costs in some countries overseas are so low that it is often more profitable to import agricultural goods than to produce them ourselves. The agricultural produce of small-scale farmers has neither the required quantity nor the quality to meet international market demand.

The analysis of renewable natural resources should be carried out in steps. The first is upstream analysis, that deals mainly with land allocation. The second is downstream analysis, which is mainly concerned with land management for allocated uses. Land allocation, in the form of selecting appropriate production systems and crops, is urgently needed. This will not only help farmers in Indonesia to face the coming free trade, but will promote trade between different parts of the country while minimizing competition and market surpluses, thus promoting national integration.

Whether crops are suitable for a particular area, or whether technology can be transferred between areas, can be assessed only if there is enough information concerning the requirements of the crop and the nature of the local agroecological conditions. Information about terrain, soil, hydrology and climate has been collected for many parts of Indonesia. However, the use of these data in making decisions about agricultural development is very limited.

Information about land resources is available in many government institutions, although it varies as to the level of detail and accuracy. This information should be utilized in a more systematic and practical way by advanced analytical methods. Defining agroecological zones on the basis of terrain, soil, hydrology and climate facilitates crop selection and agrotechnology transfer. This approach can also improve the efficiency of research and the potential impact of technologies generated by research.

Renewable Resources and Analysis Strategy

The Dynamic of Resources and Hierarchy of Management Domain

Renewable natural resources that directly affect agriculture can be roughly divided according to their relative stability. The most stable are terrain and soil, which probably will not change much within our lifetime, followed by aerial resources. In the field of agriculture, the dominant aerial variable will be climate. Although weather may change dynamically over a month or a season, climatic change will occur over a longer period, probably within a time frame of 30 to 50 years. Biological resources such as plants, microbes and animals are more dynamic. In the case of plant pests and diseases, populations often flucturate seasonally, depending on the availability of food. The most dynamic resource is human beings, who can move freely at any time.

Because some resources are dynamic and some are stable, inventory characterization, delineation and analysis must be carried out in a systematic way. Natural characteristics such as climate, terrain and soil should be put on one map. They must be distinguished from man-made features, which can change in a relatively short time. These include land use (such as the interaction between soil, plants and human beings) and infrastructure (such as roads, bridges and ports) that can be built or change within a year. Otherwise, we may have to prepare a new map every few years. With the development of computer technology such as geographic information systems (GIS), this information now can be put into different layers that can easily be lain one on top of another for analysis.

The ability of plants to convert solar energy, water and nutrients by photosynthesis into starches, sugars, fiber etc. is determined by the environment. The agroecological approach can accurately delineate the condition of this environment, and select a sustainable agriculture that is technically sound, economically feasible and environmentally safe. The agroecological approach sets a two-level hierarchy in resource management, each with different inputs and outputs. The first is for national level planning, with an expected output of production systems and crop choices. This level is related to resource allocation, and should be the first activity to be carried out. The end users of this output will be policy makers and planners.

Terrain information is the main factor in determining production systems. Fragile lands with steep slopes and marginal soils must be kept covered with natural vegetation. The main climatic information required is moisture and temperature regimes, to assess the best crop choices. The main soil information that limits agriculture is texture, acidity and drainage. These data can be obtained from field surveys.

For the second hierarchy, the agroecological approach is expected to provide information on crop management. This includes the selection of cultivars, cropping pattern and time of planting, as well as soil management, including fertilization and irrigation. Information about soil and climate is required to analyzed the expected outputs ( Table 1(1217)). If more detailed information is available, crop yield can be predicted and an economic analysis for particular crop and soil management can be carried out. Analysis of this kind needs information on the cost of production and the market value of the produce under different market conditions.

Climate and Land Resources As Criteria in Agroecological Zoning

Traditionally, land resource surveys have placed much more emphasis on soil than on climate. Consequently, climate resource inventories are seldom associated with soil information. However, climate and soil are closely related. Climate is an important factor in soil formation. Because soil formation is a long-term process, general climatic information on an annual basis is adequate.

However, a more detailed information is required in order to assess whether particular crops are suited to a particular climate. Monthly data over a period of 20 - 30 years are not generally considered sufficient. To predict crop performance at the field level, daily weather data are required. The time of planting strongly affects the performance and yield of annual crops. The weather variables most affecting crop growth are rainfall, maximum and minimum temperatures, and solar radiation. These data can be used in the simulation of dynamic plant growth processes such as photosynthesis and respiration.

Weather and climate information that is limited in quantity and poor in quality needs to be enriched to make it usable at the detailed planning level. Many soil surveys on a semi-detailed or detailed scale provide only general climatic information. Such limited information allows only crop suitability ratings, not the detailed modelling of crop performance. If such survey results were supplemented with adequate weather information, they could provide better insights into plant management options such as cropping and time of planting, as well as soil management options such as fertilizer inputs or the need for irrigation.

Technology recommendations appropriate at the field level can only be made if the decision-making processes of farmers, and the limits imposed on them by the availability of production resources, are well understood. As part of this understanding, the availability of a resource inventory covering both soil and climate is imperative. The Center for Soil and Agoclimate Research, established in 1905, has conducted many soil surveys, but only in 1986 did it begin to develop a soil database. This soil information is being linked to a geographic information system for easy retrieval, and for updating as more recent data are obtained. At the present moment, the database contains only land resources data for Sumatra on a reconnaissance scale. It will be an enormous task to computerize the land resource data obtained from surveys in other areas, in addition to the storage of data from on-going and future surveys.

Appropriate Land Allocation for Sustainable Development

Production Systems

Sustainable agriculture can be attained only when land is used in a proper way. When land is utilized improperly, productivity rapidly falls and the ecosystem is jeopardized. Proper land use ensures that resources can be used for future generations. Integrated land use management should be directed at an optimal, sustainable use of natural resources. This may include regulation of hydrology, climate stabilization, and preservation of biodiversity such as gene pools, wild-life and plant habitats, as well as appropriate research and education.

Proper land utilization improves efficiency in the production process, because relatively few inputs are required to attain the desired output, thus making production more profitable. Forest or permanent tree crops provide natural land cover in the upper watershed. In coastal regions, mangrove forests should also be conserved to protect the coast from erosion. In cold or dry areas, forests do not have the diversity of those found in warm humid areas. There is a growing awareness that native vegetation _ savanna, shrubs and trees _ needs to be conserved.

Gradient is a critical factor, because erosion and soil degradation are a real threat to agriculture in hilly regions. Steep slopes also limit the use of agricultural machinery and draft animals for soil tillage. In such areas, therefore, it is usual to grow perennial crops such as trees or pasture.

Energy efficiency also needs to be considered. On farms in steep slopelands, the labor required to transport agricultural inputs to the farm, and produce to market, may be very expensive. Labor-intensive agriculture on sloping lands is unlikely to be feasible if labor costs are relatively high.

High-value horticultural crops such as temperate flowers and vegetables are often cultivated on terraced fields in mountainous regions, to take advantage of the cooler temperatures at high altitudes. However, terracing is not feasible on all soils. Soils formed from loose parent material such as sandstone are vulnerable to landslides when terraced. Terracing highly weathered acid tropical soils will expose the high-aluminum, infertile subsoil and reduce cropping options.

If the soil is suitable, annual crop agriculture is recommended when the slope is 8% or less. It is recommended on such slopes when soils are formed from quartz sand or deep peat, or are high in gravel or stones, making soil tillage difficult. Soil with grey clay close to the surface can be reclaimed only when reduced conditions can be maintained, as when it is used to grow paddy rice. Otherwise, it should be left in its natural state and used to grow useful trees such as tea-tree (Melaleuca leucadendra). Land with a gradient of 8 - 15% is recommended for agroforesty, in which annual crops are cultivated along with perennial trees. Land with a gradient of 15 - 40% should be used only for perennial crops such as fruit trees, plantations, or forest.

Selection and Management of Annual Crops

Selection of suitable crops for specific locations is based on slope, soil texture and acidity, as well as on moisture and temperature regimes. Crops that are suited to a set of environmental conditions require fewer inputs than those which are less suitable. Unsuitable crops will give lower yields of poorer quality. In the long run, such land use may not be economically feasible.

Crop growth is generally constrained by an excess or shortage of water, and by extreme temperatures. Soil constraints are usually easier and less expensive to alleviate than those of climate and moisture. Temperature and moisture regimes can be combined to classify the environment according to which crops are suitable (Eswaran 1984).

In the production of annual crops, suitable crops and cropping patterns are determined by the water supply, based on the number of wet months i.e., those in which rainfall exceeds evapotranspiration and other losses. Crop selection should include perennial crops and hedgerows used in alley cropping, and the grasses and legumes used as cover crops and for terrace stabilization. In tropical or subtropical zones with sufficient rainfall, at least three crops can usually be grown each year.

Because rice is the staple food in Asia, it should be grown at the beginning of the rainy season. The rice plants may be intercropped with maize and cassava. After harvest, a more drought-tolerant crop such as cowpea should be planted, to avoid crop failure in the case of drought. At the end of the dry season, cassava can be harvested to improve soil aeration and as an alternative means of soil tillage for the next crop.

Areas with limited moisture and no irrigation can usually be cropped at least twice a year. Rice and maize are recommended early in the rainy season, followed by secondary crops, particularly grain legumes. Grain legume crops for soils without acidity problems are normally soybean or mungbean, but peanut or cowpea are generally grown in acid soils. Peanut requires calcium, particularly during pod filling. Calcium levels are usually low in acid soils, but because it is needed as a fertilizer, rather than a soil conditioner, small amounts of either lime or gypsum are adequate.

In semi-arid moisture regimes, rain occurs in only a few months each year and only one crop is possible. For such areas, short-duration, drought-tolerant crops such as sorghum or some cultivars of maize are recommended, and also pigeon-pea (annual or perennia). The deep-rooted pigeon-pea can extract water from the relatively moist subsoil during the dry season. Some legume forage crops such as Stylosanthes are also drought tolerant.

Land Management for Specific Uses

Soil Management

Clay minerals play a key role in determining the capacity of a soil to retain nutrients and moisture. They are therefore of great importance in fertilization and irrigation management.

For this reason, recommendations on the type of fertilizers and the method of application for a particular crop are based to a large extent on soil clay mineralogy. The fate of phosphate and potassium fertilizers in the soil is strongly affected by clay minerals. Soils derived from volcanic ash contain large amounts of amorphous minerals such as allophane. These retain much of the applied P, so that it is not available to plants. The application of phosphate fertilizer to such soil in bands or hills along with organic matter will reduce phosphate retention.

Phosphate retention also occurs to a lesser degree in soils with oxidic and kaolinitic mineralogy. Growers are recom-mended to apply phosphate in the form of phosphate rocks that can provide calcium as well as low cost P on acid soils. Because these soils generally have a low cation exchange capacity, potassium fertilizer is split into several applications to prevent losses by leaching. However, in soils where the clay tends to swell when wet and shrink when dry, (smectitic), potassium fertilizer can be applied all at once before planting, because the nutrients are retained until it is absorbed by the plant.

Nutrients in coarse textured soils are prone to leaching losses. The application of organic materials such as compost to these soils will increase their nutrient retention capacity.

Tillage of soils with smectitic minerals is particularly sensitive to the water content of the soil. Such soils require higher levels of energy to till when they are dry or contain excess water. Nitrogen fertilizer is needed by most crops, although legumes need less of it. The type of nitrogen fertilizer applied is usually based on soil acidity. Ammonium sulfate and potassium sulfate are not recommended for acid soils. Urea tends to increase soil pH, and is applied more often to acid than to alkaline soils.

Sustainable Upland Farming

Whether upland farming is sustainable depends not only on the condition of the physical environment, but also on social and economic factors, infrastructure and government policy. Policy becomes more significant with the integration of the Indonesian economy into the world economy. To be sustainable, a technology must not only be technically sound, but it must also be environmentally safe and economically feasible. On other hand, however profitable an agricultural enterprise is, it cannot last long if it causes the physical environment to become degraded.

Upland agriculture on tropical soils that are generally acidic with low fertility should focus on perennial crops such as trees or pasture, rather than on traditional food crops (Amien 1990). Erosion can be reduced by maintaining a permanent soil cover. Decomposed plant residues in roots, leaves, etc., will improve the soil's physical condition and increase the percolation rate, reduce runoff and eventually diminish erosion. Organic materials also improve soil fertility, by directly adding to the soil's nutrient supply or by increasing the nutrient absorption of plants. Green manure from legume crop residues is reported to detoxify aluminum in acid soils (Hue and Amien 1989).

The continuous cultivation of annual food crops using a low level of inputs will collapse within a relatively short period, mainly due to weed infestations (Sanchez et al. 1987). This can be seen in the increasing infestation by Imperata grass (Imperata cylindrica) of land that was formerly cultivated but is now abandoned. Continuous monoculture of the same crop does not produce a sustainable yield because of the buildup of pathogens (Valverde and Bandy 1982).

Several methods have been suggested of maintaining the productivity of upland farming on acid soils. Von Uexkull (1982, 1984) suggested a low-cost management system, using a leguminous cover crop and a variant of shifting cultivation within the crop cover area. Another way of restoring and maintaining soil fertility is alley cropping. Trees or shrubs are grown in hedgerows along the contours, with strips of annual crops grown between the hedgerows. The deep-rooted, leguminous trees used in the hedgerows recycle the leached nutrients that cannot be reached by shallow-rooted food crops, as well as preventing erosion. Cuttings from the leguminous trees provide green manure, as well as fodder for livestock. Nevertheless, the long period of time between planting hedgerows and being able to prune them for fodder and manure, and the high labor demand, make this method less attractive. Furthermore, it is not yet clear whether the hedgerows can recycle leached nutrients effectively if they are frequently pruned (Amien 1990).

Agroforestry is the term given to any type of farming involving trees. Although agroforestry is widely practised by indigenous people, it is the least studied of all tropical agricultural systems. Research into agroforestry will require an interdisciplinary approach that must include agronomists, anthropologists, geographers, rural sociologists and ecologists, as well as economists. It also implies a perceptual change, to an emphasis on sustainability rather than import substitution (Hecht 1986). Basic agroforestry techniques need a lot more study, such as the best plant combinations and their spatial arrangement in different kinds of environments. Interaction between different plant species is usually site specific, making it difficult to generalize from isolated studies. In terms of the technology available today, agroforestry can be regarded as a promising field of research for the humid tropics, but not as a system which can be widely recommended for agricultural development (Alvim 1982).

Crop diversification by planting several crops in each field has many benefits. Even during the worst times, at least some plants will give a yield. There are also more crop residues to provide livestock feed, green manures and mulches. The residues of legume crops such as Leucaena and Sesbania can provide protein-rich feed supplements. The role of livestock in integrated upland farming is very important, as a source of milk and meat as well as savings and capital. Livestock also serve as draft animals for ploughing and transporting agricultural inputs and produce. If legume crop residues are used livestock feed, rather than applied as a mulch or green manure, the nitrogen and carbon cycles become more efficient.

Decision Support Tools

Expert Systems

Computer technology is now playing a major role in agrotechnology transfer and land evaluation. The role of computer technology in database management, simulation, geographic information systems and expert systems has gained wide acceptance during the last decade (Jones et al. 1986). Expert systems belong to the field of artificial intelligence. The name refers to the use of computers to solve problems in ways that would be natural to humans (Waterman 1986).

An expert system can be seen as a computing system which uses organized knowledge about a specific field to solve a problem. Expert systems can be developed for diagnosis, classification, decision-making, tutoring, retrieving information etc. Although traditional computer programming techniques have been used to solve these same kinds of problems, a major difference exist between traditional programs and expert systems. In traditional programming, the problem-solving logic and knowledge are integrated together in the program code, and are hidden from all but the programmers. In expert systems, the knowledge remains separate and easily accessible to the user, the human expert, and the programmers (Jones et al. 1986).

The most innovative part of expert systems is the ease with which a wide range of information can be presented and used in making decisions. This information ranges from quantitative information, including statistical relationships such as regression equations and physical and chemical laws, to less precise concepts and ideas that have been gained by experience in the field.

Expert systems can be connected to external databases. By having access to a database, the expert system can infer site conditions from available data by considering, for example, the geographic location of the site.

Expert Systems for Selecting Crops Production Systems in the Tropics

An expert system for which analyzes the suitability of different crops and production systems has been developed for Indonesia (Amien 1986). It tries to provide recommendations on appropriate land use based on characteristics such as slope, soil texture, acidity and drainage. If factors such as steep gradients, very coarse soil texture, deep peat, or very low pH are present as limitations, the system recommends different agricultural systems which are appropriate for such conditions.

If the crop suitability assessment mode is selected, the system will request additional data on moisture and temperature regimes. Options will be given for a wide range of cereals, root crops, grain legumes, fiber crops, oil crops, beverage crops, vegetables and fruit crops, and cash crops such as sugarcane, tobacco, rubber, and pepper, based on soil and climatic conditions of the land. If the land is suitable only for forestry, a range of timber species and other tree species is provided. Crop suitability is generally limited by inadequate or excessive water, or by extreme temperatures.

The system also suggests cropping patterns on the basis of the water supply. Although water supply data is not always accurate, it can be inferred from drainage and the number of consecutive wet months. Recommendations on methods of applying P and K fertilizer are based on soil clay mineralogy. Information on soil mineralogy is also the basis for recommending other soil management options, such as organic manuring and appropriate methods of soil tillage. The system also provides cautions about problem soils such as potential acid sulphate soil and soil with an alkaline pH.

Often the user does not have enough data to feed the system. In this case, the system tries to use information gained from the user's experience in the field. It asks simple questions on specific soil and land characteristics. Approximate soil acidity is inferred from the natural vegetation. The moisture regime is inferred from soil drainage, and from the number of consecutive dry months in which the monthly rainfall is less than 60 mm. Because the system is designed for tropical regions, the soil temperature is inferred from the height of the land above sea level. Land at an elevation of less than 750 m is in the hottest temperature class, land at 750 - 2000 m is the intermediate class, and land above 2000 m is the coolest.

It is more complicated to determine soil clay mineralogy. This is based on several soil characteristics, including the parent material, the soil texture, the color of the subsoil, and whether or not the soil cracks open during the dry season. The color of surface soil is usually darker than that of the subsoil, because of its higher organic matter content. Therefore, input requested by the system includes the color of the subsoil. If oxidic minerals are present, the color of subsoils is reddish or yellowish. Information on soil clay mineralogy is very valuable in soil management.

Ideally, an expert system should have a large, comprehensive knowledge base to provide detailed information on land use and soil management. However, a large system of this kind requires sophisticated computers. Since the end users are likely to be extension workers who have limited access to computers, the system was designed to be compact.

An Example of Land Allocation Analysis for Sumatra, Java, Kalimantan and Sulawesi


Sumatra is the second largest island of Indonesia, with almost a quarter of the country's total area. Because of its vicinity to Java, and to the capital city, Jakarta, Sumatra is the most developed region in agriculture and industry outside Java. Almost all of Sumatra is cultivated, with plantations of oilpalm, pepper, coffee, and rubber etc. A favorable climate, especially a well-distributed high rainfall, has facilitated agricultural development. Agriculture related industry flourishes, from fertilizer plants to paper mills, and from oil palm and rubber processing to pineapple canning.

Mainly on the basis of relief, since Sumatra has a rather uniform warm and humid climate throughout the year, Sumatra is divided into six regions (Scholz 1983). It is also divided into six zones, based on gradient, soil characteristics and climatic data. As recommended by the expert system, zones I, V, and VI must left in forest to protect the environment in general. Zone II can be utilized for permanent crops only, Zone III for agroforestry, and Zone IV for annual crops.

Land is designated for forestry if it has very steep slopes, or because of the poor has quality of the soil. Other forested areas are mangrove forest along the coast growing in marine soils. Suitable tree species for forest on deep peat soils include damar (Agathis sp), ramin (Gonystyllus bancanus) and kruing (Dipterocarpus sp.).

In the permanent crop zone, crops are selected according to the soil and moisture regime. Because Sumatra is almost uniformly humid, plant performance is limited mainly by the height above sea level. On lands with an elevation of up to 750 m, the crop options are rubber, coconut, oil palm, and fruit trees. In areas higher than 750 m, the most suitable crops are tea, citrus and cinnamon.

According to the forestry land use census, the total forested land in Sumatra is 55.3 million ha (BPS, 2000). However, the recommended forested area based on agroecological zoning is only about 41% of the total forest area. This implies that more than 25 million ha of the current forested land can be converted into perennial tree plantations or used for agroforestry or annual crops. However, optimal land use is hindered by the land classification system, under which 65% of the land lies within the Department of Forestry land boundaries (although about 30% of it is not under forest cover) (World Bank 1991).


The island of Java is one of the most densely populated areas in the world. A chain of volcanoes, some still active, have enriched the soil so that it is generally very fertile. Since ancient times, Java has been a center of educational, economic, cultural and political activity in Indonesia. Java has been until recently the main producer of rice and sugar in the country. The fertile plain in the north of the island is intensively cultivated throughout the year. To the south, fertile agricultural lands have been developed in several river basins.

However, because the growing population has already occupied all available land in the plains, people have invaded the hilly and mountainous regions. If these are farmed without conservation measures, the result is erosion and increased flooding. For this reason, many reservoirs have been constructed to manage water resources for irrigation and sanitation, as well as hydroelectric power. About 60% of the population of Indonesia live in Java and Madura, and almost all of the land is being utilized for agriculture (75%). With a population growth of 1.7% annually, the number of families that depend upon agriculture will increase by about 150,000 each year. This will result in the conversion of forest land at a rate of 18,000 ha annually, with about 40,000 ha of agricultural land being converted each year into residential and industrial purposes (Word Bank 1990).

Java's well-developed infrastructure and the availability of skilled human resources have made it the center for industrial expansion that has taken place on much of the fertile agricultural land.

The increasing population of Java is augmented by migration from the outer islands of people attracted by educational and economic opportunities. Three of the biggest urban centers in Indonesia are located in Java. Transmigration and family planning programs are the principal policies to overcome these problems.

About two thirds of Java is hilly or mountainous, with only one third being relatively flat. Upland agriculture differs from lowland rice farming in several ways. It is characterized by diverse crops, low productivity and low wages. Upland farming offers fewer off-farm employment opportunities than lowland rice farming, and the infrastructure such as roads and extension services is poor.

Agricultural production plays a greater role in regional development when it generates not only raw materials, but also value-added processed goods from agro-industry. With good planning, such as cultivar standardization within areas of sufficient economic size, investments into industries based on the processing of agricultural products will generate incomes and jobs.

Java is divided into six zones, each further divided into six subzones. In contrast to Sumatra, Java has no significant areas of peat soil, but has a thin strip of sand dunes on the southern coast. This strip, like some of the marine soils along the northern coast, must be utilized as a buffer forest to protect the coast from erosion. Java has more diverse temperature and moisture regimes, including some semi-arid areas.

Because of the high population density, much of the land on Java that should have been preserved for forestry and conservation purposes has already been used for agriculture. Just over 6 million ha of Java are in forest. Because of population growth, these areas will undoubtedly continue to diminish. The land currently within the forestry deparatment boundary in Java is about 22%, or about 800,000 ha less than the recommended area. Furthermore, many of the designated forestry lands are not in fact forested.

Between 1980 and 1999, the population of Java increased by 29.16 million. With the exception of tree crops, which increased by 0.004 ha, the area of agricultural land decreased. The area cultivated in annual upland crops and paddy rice fell by 0.03 million ha and 0.04 million ha, respectively. Moreover, the character of rice-growing areas has been changing. As lowland rice fields become covered with houses and factories, to compensate for these losses, 0.49 million ha of steep lands with slopes of more than 15% have become used for agriculture. Although most of this sloping land is terraced, this improper land utilization causes erosion, and land degradation, and disrupts the hydrologic function of the watershed, resulting in frequent floods and droughts.

Intensive agriculture that has expanded into steeply sloping areas often causes disasters such as landslides and floods. Agricultural systems on steep land should be based on perennial crops so that the soil is protected by the crop's canopy and rooting system. (Erosion is caused mainly by the direct impact of raindrops onto the soil surface), moreover, the rooting system of trees helps to prevent landslides. By selecting high-value crops and by the development of agro-industry, the well-being of the people can be improved in a long-term and sustainable way.

The best way to protect sloping land from erosion is to maintain a continuous vegetation cover, particularly during the rainy season. Other than perennial trees, many forage grasses also serve to prevent erosion, because of their rooting systems and dense plant cover at the soil surface. In many studies, grass strips have proved to be more effective in preventing erosion than tree crops. Because of their relatively low canopy, grasses do not compete with food crops for sunlight in multiple cropping systems. Many upland plots in Java have been terraced. The terraces are more stable if grass strips are planted along the edges and banks.


Kalimantan is a part of Borneo that consists of four large provinces. Together, they make up 28% of Indonesia's total land area. Kalimantan can be divided into hilly and mountainous regions with gradients of more than 15% (about 30 million ha); flat wetland (about 8.7 million ha), of which about half have peat soils; undulating lands with gradients of 5 - 8 % (about 6.1 million ha), with the remainder made up of dry plains and rolling hills. Crossed by the equator, Kalimantan has two periods of high rainfall each year. The climate is warm and humid, with no distinct dry season.

Although Kalimantan has abundant water resources, the poor soil and steep gradients limit its utilization for agriculture. Kalimantan is one of the least populated islands in Indonesia, with only 20 persons per km2, compared to 945 persons in Java and 88 persons in Sumatra. Because of the scarcity of labor, the main agricultural products from Kalimantan are those which have the lowest labor demand, such as timber, rubber, pepper and oil palm. However, Kalimantan is rich in energy and mineral resources, including oil, gas and coal.

Forest cover needs to be preserved on more than 51% of the Kalimantan's to upstream watersheds, and about 14% of downstream areas. At present, 3.7 million ha are used for crops, including 1.1 million ha for lowland crops, and 1.7 million ha for upland agriculture (BPS 2000), the rest being used for agroforestry and plantatio

ns. According to agroecological conditions, this agricultural area could be extended. Kalimantan could support 7.7 million ha of plantations, 4.4 million ha of lowland agriculture and 4.7 million ha of upland crops. Optimizing land use needs to be carefully planned, to avoid failure. In view of the poor infrastructure and unfavorable soil conditions, as well as the labor scarcity, the emphasis should be on less labor-intensive agriculture such as tree plantations.


Covering more than 18 million ha, Sulawesi is a rugged island with many active and extinct volcanoes. About 60% of Sulawesi is mountainous, and 14% is hilly. Volcanic activity has enriched the soil, and the fertile volcanic soils of northern Sulawesi are Indonesia's main producers of coconuts and cloves, while those of South Sulawesi are important for rice. Rainfall patterns vary, with some areas having a single rainy season and others having two.

Sulawesi is relatively lightly populated, with only 75 persons per km2. The population is denser in the fertile volcanic soils of the north and south. Well over half the population earns a living from agriculture. With a high proportion of steeply sloping land, about 83% of Sulawesi (14.9 million ha) should be left forested for conservation purposes. The area suitable for agriculture is relatively small, only 3.5% for tree plantations and agro-forestry, 9.2% for irrigated agriculture and 3.9% for upland agriculture.

At present, a larger area than this is being utilized for agriculture: 22.09 million ha of tree plantation, 1.74 million ha of upland annual crops and 0.89 million ha for of paddy fields. This indicates that more than 1.5 million ha which are supposed to be kept as conservation areas are being used for agriculture. Furthermore, from 1979 to 1999 tree plantations increased by 1.25 million ha, compared to only 0.25 million ha for paddy rice and 31 thousand ha for upland annual crops. The area recommended for tree plantation and agroforestry is only 0.64 million ha, but tree plantations, mainly of coconut and oilpalm, have been planted in many areas suitable for annual crops. Understandably, because of the limited labor available, farmers have opted for farming systems with a low labor requirement, such as tree plantations.

Databases and Gis

Considering the limited quality and quantity of existing data, this first approximation of the agroecological zones of Sumatra, Java, Kalimantan and Sulawesi will be refined and improved as more data becomes available. Improvements will be facilitated by storing basic data and analytical results on the database, along with digitized maps in the geographic information system. At a scale of 1:1,000,000, the map of agroecological zones is appropriate as a basis for national development planning, so that the database described here has been developed on the more pragmatic basis of a "minimum database".

The database contains information about terrain, soil and climate on a mapping unit polygon basis that consists of slope, soil texture, acidity, drainage, elevation, tem-perature regime, moisture regime and soil suborder based on Soil Taxonomy. The database also contains administrative locations and a range of agricultural commodities that are suitable for each area. These commodities cover various cereals, root crops, grain legumes, oil, fiber and beverage crops, vegetables and fruit crops, as well as cash crops such as sugarcane, tobacco, rubber and pepper.

Area Delineation for Land Management Research

Human beings have intervened to shape the earth's surface in both positive and negative ways. Driven by the basic need for food, people have for thousands of years tried to produce their own food and clothing from the land they occupied. This is still happening in many places in the outer islands of Indonesia, where traditional shifting cultivation is being practiced. The present status of the land in many parts of the outer islands, where more and more land is being classified as a forest conservation area, does not give the farmers any choice but to utilize the land by shifting cultivation. In more developed parts of the country such as Java, the land status inherited from colonial times, when upland agriculture was practiced on high, steep slopes, does not favor a more rational utilization of lands.

Because different regions are at different stages of development, the present land use is not always rational or appropriate in terms of environmental sustainability and economic feasibility. If the recommended land use based on agroecological conditions is compared to the actual land use, large areas can be seen to be over-utilized or under-utilized. By comparing the ecologically optimum land use and actual, present land use, the types of interventions, along with research programs to support the intervention, can be properly formulated.

In areas where land is already properly utilized, the aim will be intensification in the form of improved cultural techniques, new high-yielding cultivars, better fertilization and pest and disease management, and improve-ments in postharvest technology and marketing. In areas where land is over-utilized, as when land appropriate for permanent crops is planted with annual crops, the intervention is land rehabilitation through diversification. This will include introducing conservation farming by planting more tropical fruits instead of the traditional annual food crops. In Java, Bali and Lombok, where more off-farm employment opportunities are available, there is often a shortage of farm labor. This shortage can be overcome by promoting commodities that require less labor such as tree plantations.

If land is being underutilized, intervention should take the form of agricultural expansion. There are many parts in the outer islands where land suitable for agriculture is classified as forest. Such land requires a change in status, and new infrastructure such as roads, bridges and ports.

For each type of intervention, different economic, social and cultural considerations have to be taken into account. By collecting this kind of information, the cost of producing particular commodities and the price of the produce can be assessed. An economic evaluation of the most profitable choice of all the suitable crops can then be made, using linear programming software.

It is necessary to run different marketing scenarios, whether at a local, regional, national or global level. Intensification and expansion are required if the market demand is high, and diversification into more profitable crops if there is an excess supply.

Capacity Building and Dissemination

Given Indonesia's vast and diverse resources, a centralized approach to resource inventory and analysis is not feasible. Both agricultural research program and training have been given a regional emphasis since 1996. Between 1996 and 1999, more than 100 people have been trained in characterizing and analyzing agricultural resources, using the agroecological approach. With guidance and supervision, these trained personnel utilized existing information to delineate regional ( Fig. 1(1000) and Fig. 2(1099)) resources, at a scale larger than 1:250,000. Such a scale provides more information, such as the delineation of peat and coastal areas into conservation and agricultural uses.

Shallow and more decomposed peats are suitable for annual crops, while other peat areas should be used for perennial crops. The area at the center of the peat dome should be left untouched, for conservation purposes. Coastal areas should maintain a buffer zone for coastal protection, by keeping the mangrove forests while allowing some aquaculture based on brackish water fishes.

As such information becomes available, sites for research assessment can be properly selected. For dissemination purposes, a series of workshops were held in 1999 and 2000. They were attended by representatives from regional development boards, agricultural services and local universities. The workshop Proceedings were published and distributed to all the stake-holders in the regions.


  • Alvim, P de T. 1982. A perspective appraisal of perennial crops in the Amazon Basin. In: Amazonia: Agriculture and Land Use Research, S.B. Hetch (Ed.). CIAT, Cali, Colombia, pp. 311-328.
  • Amien, L.I. 1986. Expert system for crops suitability and agriculture systems in the tropics. Indonesia Ag. Research and Development Jour. 8 (3 and 4): 72-75.
  • Amien, L.I. 1990. Utilization of Acid Tropical Soils for Sustainable Agriculture. Indonesia Ag. Research and Development Jour. 12, 2: 17-22.
  • Amien, I. 1997. Land resources analysis for agriculture development. Indonesia Ag. Research and Development Jour. 19 (1 and 2): 1-12.
  • Amien, I. 1998. An agroecologycal approach to sustainable development. In: Multiple Objectives Decision Making for Land, Water and Environmental Management, S.A. El-Swaify and D.S. Yakowitz (Eds.). Lewis Publisher, Boca Raton, Florida, USA, pp. 465-480.
  • BPS. 2000. Statistik Indonesia. Biro Pusat Statistik, Jakarta.
  • Eswaran, H. 1984. Uses of soil taxonomy in identifying soil related potentials and constraints for agriculture. In: Ecology and Management of Problem Soils in Asia. Food and Fertilizer Technology Center, Taipei, Taiwan.
  • Hue, N.V. and I. Amien. 1989. Aluminum detoxification with green manure. Comm. in Soil Sci. and Plant Anal. 20 (15 and 16): 1499-1511.
  • Jones, J.W., H. Beck, R.M. Peart and P. Jones. 1986. Application of expert system concepts to agrotechnology transfer. Proc. International Soil Sci. Congress, Hamburg, Germany.
  • Sanchez, P.A., J. Benites, and D.E. Bandy. 1987. Low-input system and managed fallows for acid soils in the humid tropics. In: Soil Management under Humid Conditions in Asia and the Pacific, M. Latham (Ed.). IBSRAM Proc. No. 5., Bangkok, Thailand pp 353-360.
  • Scholz, U. 1983. The Regions of Sumatra and their Agricultural Production Pattern. SARIF, West Sumatra, Indonesia.
  • Valverde, C.S. and D.E. Bandy. 1982. Production of annual crops in the Amazon. In: Amazonia: Agriculture and Land Use Research, S.B. Hecht (Ed.). CIAT, Cali, Colombia, pp. 243-280.
  • Von Uexkull, H.R. 1982. Problem soils for food crops in the humid tropics. Proc. Int. Symp. Distribution, Characteristics and Utilization of Problem Soils. Trop. Agricultural Research Centre, Ministry of Agriculture, Forestry and Fisheries, Japan. Tropical Agriculture Research Series No. 15, pp. 139-153.
  • Von Uexkull, H.R. 1984. Managing acrisols in the humid tropics. In: Ecology and Management of Problem Soils in Asia. Food and Fertilizer Technology Center, Taiwan, pp. 382-386.
  • Waterman, D.A. 1986. A Guide to Expert Systems. Addison Wesley Publishing Company, Reading, MA, USA.
  • World Bank. 1990. Indonesia: Sustainable Development of Forests, Land and Water. The World Bank, Washington, D.C.
  • Yost, R.S., S. Itoga, Z.C. Li and P. Kilham. 1986. Soil acidity management with expert system. Proceedings, IBSRAM Seminar on Soil Acidity, Land Clearing and Vertisols. October 13-20, 1986 Khon Kaen, Thailand

Index of Images

Figure 1 Agroecological Map of Southern Sumatra for National Planning (Original Scale 1:1,000,000)

Figure 1 Agroecological Map of Southern Sumatra for National Planning (Original Scale 1:1,000,000)

Figure 2 Agroecological Map of Southern Tip of Sumatra for Regional Planning (Original Scale 1:250,000)<BR>

Figure 2 Agroecological Map of Southern Tip of Sumatra for Regional Planning (Original Scale 1:250,000)

Table 1 Expected Outputs and Minimum Data Requirement at Different Scales of Land Resource Information

Table 1 Expected Outputs and Minimum Data Requirement at Different Scales of Land Resource Information

Download the PDF. of this document(706), 490,860 bytes (479 KB).