Recouping With Rice

The seemingly endless pale-green plains of the South American savannas have traditionally been devoted to cattle ranching. Many farms span more than 200 hectares. Aside from a few grass species, the acid soils of the savanna regions inhibit most plant growth. Many parts of this vast terrain have become degraded from overgrazing and the ensuing soil erosion. Now, however, ranchers are cultivating and savanna soils are becoming productive thanks to new rice varieties.

The International Center for Tropical Agriculture (CIAT), with support from the Inter-American Development Bank, has successfully developed a rice-pastures system where new rice varieties are cultivated alongside selected pasture grasses and legumes. Their research has brought the opportunity of making the savannas, all 243 million hectares of them, greener and healthier.

Soil Solutions

Past attempts to upgrade savanna pastures for livestock by sowing improved grasses led to declining soil fertility from soil compaction, erosion and loss of soil organic matter. Legumes were combined with the grasses, adding nitrogen to the soil and improving its overall quality. However, the start-up costs made the plan impractical for small farmers. To make the investment worthwhile, the farmers needed to be able to harvest a crop as well.

On-farm trials were conducted on the Matazul farm in Llanos Orientales, Columbia, from 1989 to 1993 as part of CIAT's long-term rice-pasture systems project. The site is characterized as having well-drained and flat lands with slopes <8%, and haplustox soils of intermediate to heavy texture. The experiment was specifically designed with plots large enough to permit grazing and the use of conventional machinery.

In 1989, two trials were conducted in 1 ha. plots with 3 replications, using the rice variety Line 3. The rice was planted in rows, and a mixture of grass/legume seed was broadcast over the planted crop. The grass/legume seed combination for the first plot consisted of Andropogon gayanus with Stylosanthes capitata and for the second plot was Brachiaria dictyoneura with Centrosema acutifolium.

The crops yielded, on the average, 2 t/ha with no significant differences between treatments. Establishment of the interseeded pasture species was excellent. Cattle weight gains during the first two years were higher than those recorded for similar pastures established with traditional methods. According to results from other trials in Carimagua, Meta, Columbia, an animal gains an average of 95 kg per year on a hectare of native savanna, but can gain 125 kg in improved grass pastures and 174 kg in grass-legume pastures. Animal performance declined in 1992, when the legumes were lost in both associated pastures and needed to be reseeded.

At the end of 1992, after one rice crop and three years of grazing, soil nutrient contents were comparable to, or slightly higher than, the original levels found in the native savanna soil (Table 1). This suggests that no degradation of the existing soil chemical properties occurred under the studied systems.

In 1993, the experiment was repeated on the same plots. Areas of 0.5 hectares were sown again with rice, using Line 3 in half the sown area and cv. Oryzica sabana 6, a new variety, in the remainder. In addition, a parallel small-plot experiment was established with the same treatments on native savanna.

The 1993 replication of the 1989 experiment significantly outyielded the results observed in 1989. Yields for Line 3 and Oryzica sabana 6 were approximately 165% and 183% of 1989 results. Yields of paddy rice of Line 3 were 2.9 and 3.7 t/ha after Brachiaria dictyoneura/Centrosema acutifolium and Andropogon gayanus/Stylosanthes capitata and for O. sabana 6 were 3.4 and 3.9 respectively. In the small plot experiment on native savanna, the highest yields were 3.1 from Line 3 and 2.7 t/ha from O. sabana 6. These results clearly show the benefits of three years of grazed-legume pastures.

To the Point

CIAT's rice-pasture system has opened up the possibility of cultivating one of the few remaining areas in the world that can be expanded for agriculture, while introducing a method that CIAT says could curtail deforestation by reducing the need for more land. However, as John Madeley states, some of the additional meat, milk and rice produced by the large estate farms will undoubtedly be exported, leaving the local markets still needy for produce. Land reform is slow in coming, leaving the many large estates in control of the majority of land. For farmers with small parcels of land in the savannas (100-200 ha.), there is now the chance to keep more cattle and cultivate a crop, but it is far from certain whether the poor and landless in South America will benefit from this new system. John Madeley. 1993. Raising Rice in the Savannas. New Scientist. p. 37-39.


Jose Ignacio Sanz
Apdo. Aéreo 6713, CIAT
E-Mail: CGI099 (CGNET)
Fax: (57-23) 6647243