Session 7
“Successes and failures: Stakeholders and
development agencies perspectives in enhancing
the livelihoods of communities on light
textured sandy soil”

The “Dr soils” program of the Land Development Department,
Thailand

Chinabut, N.

Office of science for Land Development, Land Development Department,
Bangkok 10900, THAILAND

Crop and soil management under integrated farming systems in Northeast
Thailand: an individual farmers experience

Polthanee, A.

Associated Professor, Department of Agronomy, Faculty of Agriculture,
Khon Kaen University

Sustainable intensification of crop-livestock systems on sandy soils of Latin
America: trade-offs between production and conservation

Ayarza, M.¹; F. Raucher²; L. Vilela³; E. Amezquita¹; E. Barrios¹;
M. Rondon¹ and I. Rao¹

Keywords: legume pastures, Cerrados, productivity livestock

1. Introduction

Sandy soils are considered marginal for agricultural activities. Many studies have shown the complexity of soil fertility problems and the challenges in developing sustainable management options, particularly on small holders farms. Because of the limited capacity of building soil organic matter (SOM) and hence nutrient stocks, farmers have to rely heavily on the use of external inputs on a seasonal basis (Giller et al., 1997). However, most of the smallholder farmers use sub-optimal amounts of fertilizer due to cash limitations and poor access to fertilizer markets. Large-scale farmers also face the challenge of replenishing nutrients that were removed from the system by both leaching losses and crop extraction. This calls for increased efficiency in use and recycling of both exogenous and endogenous nutrient pools in the cropping systems (Mapfumo and Mtambanengwe, 2004). In practice, soil conservation and crop production technologies for sandy soils have resulted in a wide range of impacts, not all favourable from an economic and environmental point of view, thus, in many cases resulting in little adoption of resource management technologies (Smith et al., 1999).

The objective of this paper is to highlight the factors driving the decisions made by farmers to intensify production of crop-livestock systems on sandy soils that rely on high or low inputs. Two case studies from Brazil were selected to illustrate the process. The conclusions presented in this paper are an attempt to generalize observations and opinions given by farmers about constraints and potential for adoption of the improved management systems.

2. Geographical distribution of sandy soils in Latin America

Extensive areas of the tropical savannas in Latin America are covered by coarse-textured soils. Quartzose sands (Entisols) occupy 30 million ha (Mha) or 15.2% of the Brazilian Cerrados. Chemically, they are characterized by having a low base saturation, low carbon levels (0.5%) and low pH of <5.5 (Adamoli et al., 1986). Sandy soils in the Colombian and Venezuelan savannas occupy roughly 20% of the whole area (4 Mha) and are scattered in patches along the main physiographic land units: upper plains, rolling lands and alluvial plains (Rondón el al, 2004). Mean SOM values reported for sandy areas where agricultural intensification is occurring are: 1.6% for sandy textures of the Colombian Llanos (Hoyos et al., 2004) and 0.5-0.8% for the sandier Ultisols of the central Llanos (Hernandez and Lopez, 2002). In general, the Eastern Llanos in Venezuela have a higher proportion of sandy soils and lower C values.

3. Land use changes in Savanna and Cerrado agro-ecosystems

Before the mid-1960’s, the Brazilian Cerrado was an empty area characterized by a low population density and the presence of extensive cattle ranching systems based on native pastures and some subsistence agriculture concentrated in the most fertile soils along the rivers. During the 70’s occupation of this area was encouraged by the Brazilian government through the POLOCENTRO and PRODECER Development Projects. Thousands of farmers from the South of Brazil moved to the Cerrados and established crop and pasture systems taking advantage of the long-term loans and ample subsidies provided by the government to purchase land, fertilizers and machinery. During the period of America 1975-80 the POLOCENTRO project incorporated 2.4 million has for agricultural production and financed 3,373 projects including the development of large-scale farming systems, construction of roads and grain storage facilities and the formation of farmer Cooperatives (Cunha, 1994). All this led to the accelerated occupation of the Cerrados. Today, there are more than 60 Mha of large scale mechanized crop systems located primarily on extensive plateaus with heavier (clay) soils, extensive cattle ranches on undulating sandy areas and small dairy farms located in the downstream in riparian areas (Ayarza, et al., 1993).

The savannas of Colombia and Venezuela experienced similar changes though at a much lower scale (Rondón, et al, 2004). By the mid 50s, they were used for extensive cattle ranching, with cattle feeding on native low quality grasses. In the mid 1960’s, the Venezuelan government promoted the introduction of large-scale tree plantations (pines, eucalyptus) and some commercial crops (rice, cotton and peanuts under irrigation). More than 0.5 Mha were planted with, Pinus caribbea in the Venezuelan Eastern Llanos. In Colombia in the 1970s, research led by CIAT and the Instituto de Investigacion agropecuaria, ICA, encouraged the intensification of livestock activities through the introduction of improved grasses (mainly Brachiaria species introduced from Africa) in association with forage legumes (Arachis pintoi, Stylosanthes capitata, Desmodium ovalifolium, Centrosema acutifolium). More recently, rapid agricultural expansion is taking place in Colombia as well as livestock intensification in Venezuela. The most important land use in the Llanos of both countries is introduced grasses (5 Mha in Venezuela and 1 Mha in Colombia). However, recently areas with lower slopes (less than 3%) have been converted to mechanized cropping (sorghum, maize) taking advantage of irrigation projects developed in the region. Forest plantations have replaced the patches of deciduous forest in Venezuela. In the process, large areas of sandy soils have also been incorporated to agricultural production.

4. Problems of the expansion of agricultural activities

The accelerated expansion of agricultural activities in the Cerrados of Brazil and savannas of Colombia and Venezuela led in a short time to an impressive increase in the export of grain, meet and milk from the region. Unfortunately, it also contributed to serious environmental problems that are associated with soil degradation, water losses and contamination and increasing costs to control pests and diseases. Intensive tillage operations in continuous cropping systems resulted in serious soil loss of physical stability and a rapid SOM mineralization in sandy and loamy soils (Lopes et al., 2004; Amézquita et al., 2002; Silva et al., 1994). Inadequate pasture management resulted in rapid pasture degradation. As much as 50 Mha in the cerrados are at some level of pasture degradation (Macedo, 1995). The degradation process was greater on sandy soils.

Individual farmers are confronting the challenge of maintaining/improving the profitability of their production systems while, at the same time, responding to the increasing pressure of society to minimize pollution and improve the quality of the soil and the water. Funding agencies and development banks are providing greater support to projects with focus on environmental protection rather than increasing agricultural productivity. However, the challenge is to optimize productivity for greater environmental sustainability.

5. Description of the case studies

Most activities reported in this paper were carried out on farms around the Municipalities of Uberlândia and Prata, State of Minas Gerais, Brazil. This region has undergone rapid intensification of land use in recent years (Oliveira Schneider, 1996). Although most of the soils are deep and well structured Red-Yellow and Dark-Red Latosols (Anionic Acrustox and Typic Haplustox, respectively, according to the USDA classification system), there are extensive areas dominated by coarse-texture soils with low natural fertility and high susceptibility to erosion. Average annual rainfall is 1,600 mm, concentrated between November and March. The dry season, from June to September, is very marked, with relative humidity dropping to less than 15%.

The work focused on developing farmer-led prototypes of two major production systems: (i) agro-pastoral systems with high inputs, and (ii) legume-based pasture systems for dairy production with low inputs for sandy and clayey soils. For the purpose of this work we will concentrate only on results from sandy soils. Previous research indicated that there are numerous advantages of the integration of crops and pastures and in the use of forage legumes to increase the sustainability of production systems (Spain et al., Boddey et al., 1996; McCown et al., 1993; Thomas et al., 1995).

Case 1: Agropastoral system for high input systems

A monitoring work was conducted in the Santa Terezinha farm located 30 km from Uberlândia. The farm has 1,000 ha and the soils are very sandy (Table 1). A team of researchers from EMBRAPA and CIAT worked in close collaboration with the owner of the farm to organize, and synthesize historical information about land use changes over time and space with the introduction of crops and improved pastures for a period of 17 years (1979-1996). The exercise allowed assessing the economic efficiency and environmental impact of the introduced systems or changes in land use pattern. It was also a good opportunity to capture drivers behind farmer’s decisions for change.

Because of its inherently low fertility and productivity potential the owner decided in 1979 to establish an extensive livestock system. After few years, he noticed that pasture productivity was declining sharply and consequently the capacity of the farm to maintain the same herd size. This crisis coincided with the return of his son from school after graduating in agronomy. Both of them decided then to plant crops to reclaim degraded pastures. With the introduction of agricultural activity on the Santa Terezinha Farm in 1983, the original system of calf fattening was transformed into an integrated system, in which cycles of crops and pastures were alternated over time and space. By 1992, all the original pastures of B. decumbens cv. Basilisk at some state of degradation had been replaced by pastures of P. maximum, planted simultaneously with maize after a 3 to 4-year cycle of cropping. From this time onward, the proportion of area under pastures remained at 40% of the total farm area (Figure 1). Despite the reduction in pasture area, the size of the herd was maintained almost constant, resulting in twofold increase in the number of calves per hectare and a 60% increase in the gross margin income as compared to the traditional systems (Table 2).

Table 1. Chemical properties of the Fazenda Santa Terezhina near Uberlandia, M.G., Brazil (mean of thirty soil samples)

The new production system also improved soil quality. During the crop cycle, soil fertility increased as a result of using fertilizers and amendments. During the pasture cycle, soil aggregation recovered, and SOM increased by 30%, compared with areas that had been planted to crops for 4 years (Ayarza et al., 2004). Lilienfein (1996) found that C, N, and P contents of the macro-aggregates were enriched during the pasture cycles.

Figure 1. Dynamics of land use change in the Santa Terezinha farm as a result of the integration of crops and pastures over time and space

Table 2. Economic efficiency of calf production in three production systems with different degrees of inten­sification

The control of pastures for the planting of crops was initially achieved by conventional tillage, at the end of the rainy season. With the progress made on no-till systems the farmer decided to plant crops on pasture residues chemically controlled. The main goal of the farmer was to take advantage of the improved soil fertility and minimize tillage operations.

Case 2: Legume based-pasture systems with low inputs

The work was conducted in dairy farms near the Municipio of Prata, 30 km west from Uberlândia. This region is located at the lower part of the watershed formed by the Paranaiba and Rio Grande rivers. Most of the region has sandy soils with low pH (<5.5) and low availability of P. Relief is gentle in the top of the “Chapadas”, however, it is pronounced toward many streams and rivers present in the region. Cultivated pastures have replaced most of the native Cerrado vegetation and now they account for 64% of total area (0.32 Mha). More than half of the 24,000 inhabitants of the Municipality in 1964 were devoted to produce milk. Milk was collected and commercialized through the Cooperative of Producers of Prata “CORPRATA”. This is a farmer-led organization created to provide credit, technical assistance and inputs to 554 associates.

Most of milk produced in the wet season was obtained on degraded Brachairia pastures. During the dry season, milking cows were fed with supplements based on cutting grass, sugarcane and concentrate. Farmers interviewed during the work mentioned that their priority was to increase milk production and reduce cost in the use of concentrate particularly during the dry season.

In order to improve performance of dairy systems and income of farmers of the region, the EMBRAPA-Cerrados Center and the Tropical Lowlands Project of CIAT developed a collaborative project between 1996-1998 to test the potential of the forage legume Stylosanthes cv. Mineirao to increase milk production during the dry season. Previous research showed that this legume is well adapted to low soil fertility soils and is capable to remain green during long dry season (EMBRAPA, 1993; Barcellos and Vilela, 1994; Ramos et al., 1996). One ha paddocks of Stylo + Brachaiaria grass pasture were established in twelve farms varying in size, herd size and productivity per cow (Table 3). A common problem in all farms except one was very low level of milk productivity per cow (3.8 l/day on average).

Table 3. Characteristics of the farms selected for the evaluation of stylosanthes guianensis cv. Mineirao

Daily production of two milking cows grazing the grass + legume paddocks was recorded during four periods of eight days each during the dry season of 1997. Production was compared to that obtained from the same cows for a similar period of time just before they started grazing the legume.

Stylosanthes cv. Mineirao had a positive effect on milk production (Table 4). Impact was greater with cows having lower productivity potential. These types of animals are common in dual-purpose systems in small farms. None of the farmers participating in the work was ready to substitute the concentrate by the use of Stylosanthes. However, during the second year some of them started reducing by half the use of concentrate without losing milk production. Others decided to keep using Stylo as a complement of the animal diet. In both cases there was a significant effect on production costs.

Table 4. Milk production of cows grazing Stylosanthes Mineirao during the dry season in small dairy systems in several farms in Prata. M.G., Brazil (mean of four periods of eight days)

Farmers also saw additional benefits of grazing the legume during the wet season since the recovery of lactating cows and their calves was faster as compared to the traditional degraded pastures. Milk increments were modest but significant for small farmers. Larger-scale farmers did not put much importance on milk output but on the benefit of grazing pastures during the dry season and the improved conditions of the pastures during the wet season.

6. General discussion

Crop-pasture integration

The integration of crop and livestock activities is relatively new to cerrado farmers but it seems to be the best option for sustainable management of sandy and clayey soils. In the case of the Fazenda Santa Therezinha the owner perceived the economic and environmental advantages of this technology and accepted the challenge to adjust his traditional production system to an innovative one with the support of his son who was returning from school with a B Sc degree in Agronomy. The introduction of crops such as soybeans and maize allowed him to improve the short-term profitability of the production system while the increased pasture productivity improved even further the long-term profitability of the livestock component. Good and stable prices of commodities in the market and credit to purchase machinery and inputs were major incentives for the change. The results of monitoring crop-livestock systems confirmed the beneficial effect of integration on the system’s productivity and on soil quality.

Despite its great potential, only a few farmers have adopted this system. This is partly because changes are needed in infrastructure and management to handle both activities. In addition, a change in mentality is needed from both grain farmer and livestock owner, who are accustomed to working with just one activity (Spain et al., 1996).

Apparently, this change is now beginning to take place among grain farmers in the Cerrados, as they are beginning to perceive the economic benefit of crop/ pasture integration, even when both activities are carried out in separate areas. Smith el al. (1999) conducted a survey to assess the factors driving land use change in the Cerrados near Uberlândia. They found that about half of the farmers interviewed, had dairy and beef cattle. During the rainy season, the animals remain in the cultivated pastures. During the dry season, the animals are confined and supplemented with silage and feed concentrates, usually produced from crops on the farm. This type of integration has made it possible to take advantage of areas unsuitable for agriculture and increase the farmers’ income through milk and meat production. In this case, farmers perceived livestock as complementing rather than substituting their principal activity of grain production.

During the last 10 years, the system is expanding to remote areas in the North of the Cerrados. Framers are using this technology to reclaim degraded pastures and improve economic efficiency of the system.

Legume based pastures

The introduction of Stylosanthes cv. Mineirao in low input systems of small dairy systems is a good example of an innovation that can generate benefits with respect to productivity and soil quality in both the short and long term, without causing major structural changes to the production system. This technology requires only small quantities of seed to establish (0.7-1.5 kg/ha), does not need additional soil pre­paration tasks, and requires very few inputs (Ayarza et al., 2004). These characteristics and the capacity of cv. Minerirao to provide green forage during the dry season give this legume clear advantages to improve the productivity of the system, during critical times. In addition to improving milk production and, the availability of N in the soil-plant system increases and permits greater stability of pastures over a longer term. Stylosanthes is very sensitive to initial competition and therefore it is important to control regrowth of com­panion grasses through proper grazing management to have a good establishment of the legume. Input of researchers to solve this problem was important.

Participating farmers were most of the time impressed by the outstanding performance of the legume during the dry season. They perceived the superior forage availability and the better condition of animals grazing the cv. Minerirao plots. However, they were cautious to move to this technology because of economic limitations, lack of knowledge about its management and availability of adequate seed in the market. After they perceived the benefits during the first year they were enthusiastic about having more cv. Mineirao in their farms.

Although technologies supported by the cooperative (sugarcane, urea and concentrate) helped to increase milk production during the dry season their cost was equal or even higher than the returns obtained by small farmer by selling the milk produced. For this reason they were enthusiastic about using cv. Mineirao. However, none of the farmers participating in the work substituted the use of concentrate during the first year of evaluation. Rather, they were interest to prove that cv. Mineirao could increase even further milk production. In the second year some of them started to reduce the use of concentrate.

The major challenge for researchers working in the project was to convince farmers that the legume would be a cheaper solution than the use of supplement and concentrate. The only way to overcome this problem was by involving, as much as possible, farmers in the evaluation process. Farmers had enough freedom to select the areas they wanted to have the legume, to participate in the evaluation process and even quit if they loose interest. On the other hand, the Cooperative requested for a continuous process of capacity building to provide technical assistance once the project was over.

Constraints for adoption

As with all new technologies and management options there are some difficulties and obstacles to their widespread adoption. Below we briefly discuss these aspects.

a) Economic, social and cultural aspects

Integrated crop-livestock systems appear to be sound from the economical and ecological points of view. However, its massive adoption will require a change in the structure of the production system. It will also require incentives, credit and a “new type” of producer empowered with tools, information and managerial skills to optimize short and long-term benefits according to market demands.

Small farmers devoted to produce milk confront the problems of lack of financial resources, education and health services. Furthermore, they are pressed to improve the short-term profitability of the system to satisfy the demands of the family. All these factors restrict their capacity to adopt any technological innovation without outside support (financial, technical, etc.). Land tenure is often mentioned as a constraint for the adoption of technologies. In this case it was not so important since most farmers in the cerrados own their own land.

b) Technological aspects

The biggest limitation for the massive adoption of improved legumes such as Stylosanthes cv. Mineirao was the lack of good seed. Some of the small farmers in Prata area tried to harvest seed from their small paddocks with little success. This bottleneck is often ignored when developing crop technologies. Alliances with the private sector and farmers organization are needed to support seed production.

Lack of continuous technical assistance to ensure good establishment and management of legume-based pastures is another constraint. Legume based-pastures require a skillful management for maintaining a proper balance between grass and legume components.

c) Favourable policy environment

New technologies and management options will require more favourable policy environments to be successful. Better dialogues amongst farmers, researchers, extensions and policy makers are essential and require much more effort from the agricultural research community.

d) Lack of effective participation of farmers

Even though there has been a considerable effort for the development of better technologies in the region by various national and international institutions, the true “explosion” in technology implementation was more a result of the effort of the farmers themselves. Farmer-to-farmer communication and farmer cooperatives were the main driving forces behind the transfer and adaptation of the experiences obtained in the southern temperate region in Brazil (Landers, 1996). Farmers need to be engaged in a dialogue on how they can arrive to solutions that suit their require­ments and circumstances. An entry point for truly interdisciplinary research would be experimentation based on farmers own practices.

To better match precision of farming systems to technologies, technology design must involve the intended beneficiaries earlier. That is moving from formal to more farmer-led as soon as the comparative advantage shifts from researcher to farmer.

Main conclusions

Land use policies and land management practices are in general designed with a long-term objective, while farmers are generally more concerned with the short-term benefits for the continuing survival. Therefore it is appropriate to identify entry points where short-term benefits could be created. Resource management technologies for sandy soils should be conceived on the need to improve livelihoods of farmers, while ensuring that the environment inherited by future generations is not degraded. A favourable policy environment is needed during the transition periods of technological change.

The strategy of on-farm research make possible to incorporate the farmer early on in the generation and assessment of the new technologies. In addition, it provides an opportunity of identifying problems earlier, and of conducting research under a broad range of conditions, as in our study, for which conditions ranged from high input to low input systems and large-scale to small-scale farmers.

A better understanding of the driving forces behind land use change at the farm and other levels is required in order to ensure that new technological and management options are appropriate and are acceptable to farmers. Adoption of legumes depends on the comparative advantage over other crops and other alternatives. The market demand, price incentives, labour, pests and diseases limit scaling up and out of legume-based technologies.

Acknowledgements

We gratefully acknowledge the support of Carlos Raucher, Cristiano Hacker and small producers from the Prata region who made possible to carry out this work. We also thank to the local staff of the COPRATA Cooperative for their willingness to help us in the collection of information. Special thanks are also extended to CPAC-EMBRAPA researchers for their technical inputs during the execution of the project. Lastly, we want to express our recognition to the effort and dedication of Paulo Humberto da Costa, our field technician, in conducting the field work in Uberlandia and Prata. Without his support it would had been impossible to do the work.

References

Adamoli, J., Macedo, J., Azevedo, de L.G., Mattos, L.J. 1986. Caracterizacao da regiao dos Cerrados. In: CPAC. Solos dos Cerrados, Nobel, Sao Paulo, Brazil, 33-74.

Amézquita, E., D.K. Friesen, M. Rivera, I.M. Rao, E. Barrios, J.J. Jiménez, T. Decaëns and Thomas, R.J. Thomas. 2002. Sustainability of crop rotation and ley pasture systems on the acid-soil savannas of South America. 17^th World Congress of Soil Science, Bangkok, Thailand, 14-21 August 2002.

Ayarza, M.A., Soares, W., da Rocha, C.M., Texeira, C.M and Bahia, F. 1993. Caracterizacao dos sistemas Agrícolas e dos Problemas de sustentabilidade em Quatro Regioes do Cerrado Brasileiro. Brasilia: EMBRAPA/CPAC, 20 p.

Ayarza. M.A, L.Vilela, E.A. Pizarro and P.H. da Costa. 2004. Agropastoral systems based on multiple-purpose legumes. In: Agropastoral Systems for the Tropical Savannas of Latin America. Eds: E.P. Guimaraes, J.I. Sanz, I.M. Rao, M.C. Amezquita, E. Amezquita, R. Thomas. Cali, Colombia. Centro Internacional de Agricultura Tropical (CIAT): Empresa Brasileira de Pesquisa Agropecuaria, EMBRAPA. 185-202.

Barcellos, A.O and Vilela, L. 1994. Leguminosas forrageira tropicais: estado de arte e perpectivas futuras. In: Simposio International de Forragicultura. Anais de XXXI Reunioao Anual de Sociedades Brasiliera de Zootenia, Maringa, EDEM p. 1-56.

Bodey, R.M., Alves, BJR., Urquiaga, S. 1996. Nitrogen Cycling and sustainability of improved pastures in the Brazilian Cerrados. In: Pereira RC and Nasser, L.C (eds.). Simposio sobre o Cerrado. EMBRAPA-CPAC, Planaltina, DF, Brazil. 33-38.

Cunha, A.S., Muller, C.C., Alves E.R.A and da Silva, J.E. 1994. Uma avaliacao da sustentabilidade da agricultura nos cerrados. IPEA. Brasilia 2 v.

EMBRAPA. 1993a. Relatorio Anual. Centro de Gado de Leite. Jiz de Fora, M.G.

EMBRAPA. 1993 b. Recomendacooes para o estabelecimento e utilizacao do Stylosanthes guianensis cv. Mineirao. Comunicado Técnico No. 49. 6 p.

Guiller, K.E., Cádiz, J., Ehaloitis, C., Sácala, W.D and Mafongoya, P.L. 1997. Building soil nitrogen capital in Africa. In: Buresh, R.J., Sanchez, P.A (eds.). Replenishing Soil fertility in Africa. SSSA Special publication 51: 151-192.

Hoyos Garcés, P., E. Amézquita Collazos and D.L. Molina López. 2004. Estrategias para la construcción de capas arables productivas en dos suelos de la Altillanura Colombiana. Informe Final 2001-2003. Proyecto No. 201504060, CIAT-PRONATTA. 79 p.

Hernández-Hernández, R.M. and D. López-Hernández. 2002. El tipo de labranza como agente modificador de la materia orgánica: un modelo para suelos de sabana de los Llanos Centrales Venezolanos. INTERCIENCIA. 27(10): 529-536.

Landers. J. 2001. How and Why the Brazilian Zero Tillage Explosion Occurred. In: D.E. Stott, R.H Monthar and G.C. Steinhardt (eds.). Sustainable the Global farm. Selected paper from the 10^th International Soil Conservation Organization meeting held in May 24-29, 1999 at Purdue University and the USDA-ARS National Soil Research Laboratory. 29-39.

Lilienfein. J., Wilcke, W., Neufeldt, H., Ayarza, M. and W. Zech 1998. Land-use effects on Organic Carbon, Nitrogen and Sulphur concentrations in macroaggregates of differently textured Brazilian Oxisols. Z. Pflanzenernähr. Bodenk. 161, 165-171.

Lopes, A.S., Ayarza, M and R.J. Thomas. 2004. Managing and Conserving Acid savanna soils for Agricultural Development: Lessons from the Brazilian Cerrados. In: Agropastoral Systems for the Tropical Savannas of Latin America. Eds. E.P. Guimaraes, J.I. Sanz, I.M. Rao, M.C. Amezquita, E. Amezquita, R. Thomas. Cali, Colombia. Centro Internacional de Agricultura Tropical (CIAT): Empresa Brasileira de Pesquisa Agropecuaria, EMBRAPA, 11-41.

Macedo, J. 1995. Prospectives fir the rational use of the Brazilian Cerrados for food production. EMBRAPA-CPAC, Planaltina, DF, Brazil.

Mapfumo. P and Mtambanengwe, F. Base nutrient dynamics and productivity of sandy soils under maize-pigeon pea rotational systems in Zimbabwe. In: Bationo, A, ed. Managing Nutrient Cycles to Sustain Soil Fertility in Sub-Saharan Africa. Academy Science Publishers in association with the Tropical Soil Biology and Fertility Institute of CIAT.225-238.

McCown RL; Thiagalingan K; Price T; Carberry PC; Jones RK; Dalgleish NP; Peake DC. 1993. A legume ley system in Australia’s semi-arid tropics. In: XVII International Grassland Congress, Proceedings. Rockhampton, Australia, and Palmerston North, New Zealand. p. 2206-2008.

Oliveira Schneider M de. 1996. Bacia do rio Uberabinha: Uso agrícola do solo e meio ambiente. PhD dissertation. Universidade Estadual Paulista, São Paulo, Brazil.

Ramos, A., Pizarro, E., Ayarza, M.A., Carvalho, M and Costa, P. 1996. Avaliacao agronómica de gramíneas forrageiras consorciadas com Stylosanthes guianensis cv. Mineirao em Uberlândia, M.G. Anais da XXXIII. Reuniao da Sociedade Brasileira de Zootecnia. Fortaleza, Ceara.

Rondón, M., Acevedo, A., Hernandez, D., Rubiano, R.M., Rivera, Y. , Amezquita, E., Romero, M., Sarmiento, L., Ayarza, M., Barrios, E. and Rao, I. 2004. Carbon Sequestration Potential of the Neotropical Savannas (Llanos) of Colombia and Venezuela. In: R. Lal and J. Kimble (eds.) Carbon sequestration potential of the main biomes of Latin America. CRP Press. In review.

Ruaysoonnern, S. and Patanothai, A. 1991. Farmer’s perceptions and the adoption of sustainable land management technologies-Thailand experience. In: Evaluation for sustainable land management in the developing world. Technical papers. IBSRAM Proceedings. Volume II, 491-508.

Smith, J., Cadavid, J.V., Ayarza, M., Pimenta de Aguiar, J, and Rosa, R. 1999. Land Use Change in Soybean Production Systems in the Brazilian Savanna: The Role of Policy and Market Conditions. Journal of Sustainable Agriculture: 15-02, 95-117.

Silva, J.E. de., Lemainski, J. and Resck, D.V.S. 1994. Perdas de Mateira Orgánica e suas relacoes com capacidades de troca cationica em solos da regiao de cerrados do oeste Ballano. Revista. Brasileira da Ciencia do Solo 18: 541-547.

Spain, J.M. Ayarza MA; Vilela L. 1996. Crop-pasture rotations in the Brazilian Cerrados. In: Carvalho EF; Nassar B, eds. Biodiversity and sustainable production of food and fibers in the tropical savannas. Proceedings of the First International Symposium on Tropical Savannas, Brasília, 24-29 March, 1996. Planaltina, DF, Brazil. p. 39-45.

Thomas, R.J. 1995. Role of legumes in providing N for sustainable tropical pasture systems. Plant Soil 174: 103-118.

¹ Tropical Soil Biology and Fertility (TSBF) Institute of CIAT, A.A 6713, Cali, Colombia. E-mail: m.ayarza@cgiar. org
² Producer, Fazenda Santa Terezinha, Uberlandia, Brazil.
³ EMBRAPA-Cerrados, Planaltina, DF, Brazil.