ÐÓ°ÉÔ­´´

(see Graphic)
Crops are a rare sight on the vast savannas of eastern Colombia. Farmers who
try to cultivate the grassy plains, even the river banks, often give up in
despair. Infertile, acidic, sandy soils seem to make the land fit only for
pasture. And even then, the cattle must be well spaced to avoid damaging
this fragile soil – no more than one cow every 10 hectares is the norm.

But now the farmers have good reason to try again. Last year, a variety of
rice developed especially for the conditions grew for the first time
alongside pasture grasses and forage plants on 4000 hectares of this
unpromising land. It also improved the quality of the pasture, and the
density of cattle rose twentyfold. If this success is repeated across South
America, the breakthrough would exploit one of the last areas in the world
that can be expanded for agriculture and would provide more food for a
continent suffering considerable malnutrition.

South America has some 243 million hectares of savanna, which, though barely
half the area of land currently used for agriculture, is more than four
times the size of France and larger than the total area under cultivation in
sub-Saharan Africa. Besides Colombia’s eastern plains, Llanos Orientales,
the continent’s savanna lands cover the Brazilian Cerrados, the Venezuelan
plains and some of Argentina, Bolivia, French Guiana and Paraguay.
Typically, the acid soils are well drained but contain too little
phosphorous, calcium, magnesium and potassium, and too much aluminium.

Overcoming these drawbacks was the task of the International Centre for
Tropical Agriculture (CIAT) in Cali, a provincial city about 300 kilometres
south west of Bogota. CIAT, founded in 1967, is one of 18 aid-funded
research centres that make up the worldwide Consultative Group on
International Agricultural Research. Under the Tropical Pastures Programme,
which CIAT has run since its first day, researchers at the Colombian centre
worked at three levels to find a way of using the savannas for sustainable
pasture and crops. They looked for new varieties of pasture grasses to
improve the quality of the grazing for animals, for legumes to increase the
fertility of the soil, and for crop plants to cultivate for food.

Out of Africa

At its Carimagua research station in the heart of eastern Colombia’s
savanna lands, CIAT evaluated and selected pasture grasses of the genus
Brachiaria. Originally from Africa, these are known to adapt well to low
fertility and acid soils although they are susceptible to damage by the
larvae of froghoppers – known as spittlebugs because of the frothy fluid
the larvae secrete around themselves for protection. Among the grasses
selected were B. decembens, which offers good nutritional value, and B.
dictyoneura, which is partly resistant to frog-hoppers.

But previous attempts to upgrade savanna pastures by sowing improved grasses
have led to a decline in the soil’s fertility. So CIAT decided to combine
the grasses with legumes, whose roots harbour bacteria that absorb nitrogen
from the air and convert it into soluble nitrates, an important nutrient for
plants.

Researchers at the centre screened more than a thousand species of tropical
legumes to find ones suited to the savannas, but which would not be so
vigorous that they killed the grass. One of the most promising, Arachis
pintoi, is known as tropical white clover, although its flowers are yellow
and it is not a clover. Indigenous to Brazil, A. pintoi is a perennial
legume that produces a dense mat of horizontal stems, known as stolons,
which run along the surface of the soil and produce numerous roots and
shoots so that the plant spreads quickly. ‘A. pintoi seems to contribute
enormously to the quality of the grass and to animal production,’ says
Indupulapati Rao, a plant nutrition physiologist at CIAT.

Richard Thomas, also from the centre, specialises in the nitrogen cycle and
estimates that legumes such as A. pintoi provide between 40 and 80 kilograms
of nitrates per hectare a year. Legumes give savanna farmers two advantages:
they reduce the need for nitrate fertiliser and improve the quality of
grazing land by adding organic matter to the poor soil as their foliage dies
and rots.

Thomas says that in tests at Carimagua over a nine-year period, from 1978
to 1987, cattle on grass-legume pastures increased their weight by 50 per
cent more than those on grass-only pasture. Measurements of milk
production, carried out in wet and dry seasons and with cows at different
stages of lactation, showed that cows on grass-legume pastures produced 20
per cent more than those on grass-only pastures.

But investing in improved pasture may still be uneconomic for some farmers
if the only benefit is a more productive herd of cattle. To make the
investment worthwhile, these farmers need to be able to harvest a crop as
well. With this in mind, the Inter-American Development Bank provided CIAT
with funds to develop a crop-pasture system for the savannas. After testing
a number of crops, rice emerged ‘as a clear winner’, says Elcio Guimaraes, a
plant breeder at the centre.

One of the problems of finding a rice for savannas is that, although the
crop grows in a wide range of soils, acid soils are usually not conducive to
its growth. (Farmers have succeeded in growing rice on Brazilian savannas
but only after heavy use of chemicals, such as lime, to reduce acidity.)
CIAT, as part of the Rice Programme that it has also been running since
1967, began by selecting varieties that had been developed for upland areas
in the 1980s by the International Institute of Tropical Agriculture in
Nigeria and the Paris-based Tropical Agronomical Research Institute (IRAT).
These varieties will grow without the flooding that is necessary for
irrigated varieties grown in paddy fields, and are common in Latin America
and Africa.

Upland rices have some tolerance to soil acidity and disease, but they are
susceptible to the brown planthopper pest and are inferior in quality to the
irrigated varieties. IRAT has developed upland varieties for Madagascar that
produce a better quality of rice and have some resistance to the
planthopper, but they are sensitive to the acidity of the soil and
susceptible to the fungus, rice blast, which is one of the most serious rice
diseases.

So CIAT crossbred the most promising varieties and, in the late 1980s,
developed rice types that are partly resistant to the planthopper.
Eventually the research led to the development of Oryzica sabana 6, which
was released to farmers in late 1991 by a government agency, the Colombian
Agricultural and Livestock Institute. In eastern Colombia last year,
farmers turned over about 4000 hectares of their savanna – between 300 to
600 metres above sea level – to rice-pastures. Researchers say the
first-year results are highly encouraging.

Under the system, farmers plant their rice and grass-legume pasture at the
same time. In May, after preparing and fertilising the land, they plant the
rice in rows and scatter the mixture of improved grasses and legume seeds.
Last year, farmers harvested the rice after three to four months and then
grazed cattle on the pasture. Adequate rainfall, equivalent to about 1500
millimetres a year, made irrigation unnecessary, and the rice yielded about
3 tonnes per hectare – three times the normal yield for an upland rice.
CIAT estimates that farmers need to obtain 2 tonnes of rice a hectare in
the first year to recover the initial costs of establishing a rice-pasture
system.

Some farmers are seeing the rice-pasture system chiefly as a means to raise
livestock production. ‘We have planted 450 hectares of rice and Brachiaria
dictyoneura,’ said an assistant on a cattle-raising farm in eastern
Colombia, ‘but intend to remain cattle raisers.’ The farm had adopted the
new system because it wanted ‘higher quality pastures to grow more and
better cattle on less land’.

Farmers found that the pasture grew faster because of the small quantities
of fertiliser they used for the rice. Normally it takes about a year to
establish improved pasture but the period reduces to four or five months
when rice is planted. The pasture is also richer, allowing the density of
cattle to be increased. Farmers who could once stock their land with only
one cow every 10 hectares are now thinking in terms of two animals per
hectare. They also found that cattle put on weight even faster when grazed
on land where rice has just grown, because of the richer grass. Under the
grass-legume regime, farmers waited two years before taking their cattle to
market; now they may be able to sell the animals within 16 months.

Quality crop

The only drawback for the farmers who grew rice in 1992 was its quality. O.
sabana 6 is considered equivalent to upland rice in quality, which sells for
about 30 per cent less than irrigated rice. The Rice Programme has recently
developed a new variety for the savannas, CT 10037, which the centre will
release to farmers this year. The new variety yields 10 per cent more rice
than O. sabana 6, and gives a grain quality that should fetch around the
same price as irrigated rice.

CIAT is keen to emphasise that rice will not succeed in poor savanna soils
if it is grown alone. Tests by the centre with monocropping, where no
grass-legume pastures were sown, show that rice yields are good in the first
year on savannas, drop considerably in the second year, and are down to
almost nothing in the third. Even when a grass-legume pasture is sown, rice
should not be grown continuously: ‘No more than three or four years,’ says
Jose Ignacio Sanz, coordinator of CIAT’s rice-pasture project. He stresses,
however, the variability of the time limit: ‘It all depends on management –
land preparation, fertilisation, weed infestation, pest incidence and so
on.’ Farmers might think in terms of having up to half their land under
rice-pasture at any one time and half under pasture only.

Sanz believes that the rice-pasture system is not only profitable but also
protects the environment. Using savanna in this way helps to maintain soil
fertility, as nutrient recycling is generally efficient. ‘Continuous
monocropping is avoided,’ he says, ‘and we don’t need a lot of land
preparation that leads to erosion, or heavy pesticide and fertiliser use.’
This also helps to reduce the amount of chemicals leaching into groundwater.
If agricultural productivity from savannas can be increased by this
rice-pasture system, there would also be less pressure to clear forests for
growing crops or rearing cattle.

Nonetheless, there is a danger that farmers with large holdings of savanna,
or people with money to buy land, will believe there is a killing to be made
out of growing rice on the savannas and will try to cultivate it in
mono-cropping fashion. This could ruin the already fragile savannas.
‘Farmers in the area are excited by the realisation that they can grow rice
on the savannas, but we want them to understand that the system is fragile,’
says Guimaraes.

Froghoppers and leafcutter ants pose another problem. Many Brachiaria
grasses are susceptible to attacks by froghoppers, and also to leafcutter
ants that cut and carry away pieces of vegetation to use as fertiliser for
the fungus on which they feed. One species, B. brizantha, is ‘highly
resistant to froghoppers, but more demanding in soil fertility,’ says
Stephen Lapointe, an entomologist at CIAT. Within a few years of sowing this
grass, the soil becomes very impoverished. The centre now has a major
breeding programme under way with a number of species including one from
Brazil, B. ruziziensis, which also has a reputation for good nutritional
quality.

The search for a grass that is resistant to leafcutter ants is less
advanced, but the problem can be alleviated by tilling the land thoroughly,
which destroys their nests.

Breeding for tolerance and, ultimately, resistance to pests and disease is
seen by CIAT scientists as vital in view of the price of insecticides and
their environmental side effects. The Colombian government has eliminated
subsidies for agrochemicals over the past two years, and this has encouraged
the search for natural methods of pest and disease control.

Weeds have also emerged as a problem in tests. The complex and irregular
distribution of the various species on the savannas means that chemical
control with weedkillers is not possible. The rice-pasture team is now
concentrating on methods of preparing the soil that will check the growth of
weeds.

Crop searching

The Tropical Pastures Programme is continuing its research on identifying
and developing legumes that, when grown as part of a rice-pasture system,
could help to increase the sustainable fertility of the savanna soil.
Lapointe says Stylosanthes capitata is particularly suitable, having an
‘excellent adaptation to very low fertility acid soils, particularly
light-textured soils’ such as the sandy soils of the savanna. Research is
also continuing on other crops that could grow in South American savannas,
including maize, sorghum and soya bean. ‘Savannas are very underutilised,’
says Sanz.

While CIAT’s rice-pasture system could not be applied directly to savannas
on other continents or agroecosystems because of the differences in soil
composition, the principles of crop-pasture integration and the research
behind it could be useful. Africa has tropical savannas similar to those in
South America, many of them in the dry Sahelian zone, which lies to the
south of the Sahara.

In South America itself, the rice-pasture system could help to alleviate
poverty among farmers with small holdings in the savannas. Many farms are
between 100 and 200 hectares, which means they can keep a maximum of only 20
cattle and grow no crops. With a rice-pasture system, farmers can keep more
cattle, grow and sustain a crop, and make a profit.

But the limited pace of land reform in South America means that the
continent’s vast estate farms still thrive. At least some of the rice, extra
meat and milk produced under rice-pasture, especially on large farms, is
likely to be exported and make little contribution to alleviating hunger and
poverty. For the many landless and urban poor of the continent, who cannot
afford the fruits of science, it is by no means certain that the system will
bring any benefits.

John Madeley is a freelance writer.