ALN logo; link to Arid Lands Newsletter Home page No. 48, November/December 2000
Linkages between Cultural Diversity and Biodiversity

Polycultures in the Brazilian drylands: A new version of an old tradition

by Marsha Hanzi

"[The polyculture model tested in]this project is financed by the local castor bean industry, which was interested in having a steadier supply of raw materials. These first demonstration fields have given spectacular results: one farmer harvested the equivalent of 1,400 Reais (about 700 dollars) from one hectare alone. This is about double the regional average income per hectare..."

The drylands of Brazil

(Back to top)
Although Brazil is best known for its vast rainforests, it is also home to an immense semi-arid region (900,000 square kilometers and growing!). Located in the Northeast of the country, this region, the Sertão, is composed of hundreds of micro-climates and specific ecosystems, rich in fauna and flora. It is also densely populated (10 million inhabitants) and highly degraded. The landholdings are a mixture of many smallholdings (5-30 hectares) and large cattle ranches, often owned by absentee owners. The local people are a mixture of original indigenous inhabitants with Portuguese, with a strong oral tradition that many were actually Jewish refugees fleeing from the Inquisition (hiding their identity by changing their names). Although the Indian tradition is virtually extinct in the region, the form of agriculture practiced until today by many smallholders is the traditional (and extremely destructive) slash-and-burn.

A recent survey estimated that 40% of this region is degraded with 25% in danger of desertification, but this estimate is based on meager evidence, at best. The fact is that with the arrival of "modern" farming and cattle-raising, this region has suffered intensive removal of the native vegetation, called Caatinga, exposing the shallow and fragile soils to hot constant winds and sheet erosion. The traditional farmers planted small (half-hectare) plots in the heart of the native vegetation, thus maintaining biodiversity and natural windbreaks. When the short rainy season was over, the rest of the year was spent hunting the local fauna. With the advent of tractors, however, these plots grew to 2-3 hectares or more, exposing the land, and virtually eliminating the Caatinga. With the arrival of commercial farming, some 20 years ago, when many farmers began to plant for money instead of for subsistence, the bank credit system demanded that these plots be planted with monocultures and maintained with commercially produced chemicals. This had the effect of drying the land even more (and dragging many would-be "modern" farmers to bankruptcy, since this model is much more expensive than the traditional one). Although the yearly precipitation averages continue to be more or less the same as historic values, there are now perceptible changes in the climate: the rains are coming ever more violently and unpredictably and the period of droughts is becoming longer and hotter.

The Sertão now depends on a totally unsustainable form of agriculture, based on corn and beans, which only produce well two years in five, or less. To compensate for the bad years, the farmers run cattle and goats and hunt native fauna, with which the remaining Caatinga is blessed. Unfortunately, in most regions, due to overstocking, this has affected the local ecosystem to a point where it is having difficulty regenerating even in rainy years, creating a downward spiral into desertification. Together with the growing climate instability, this situation has provoked social and economic instability, forcing millions of these dryland inhabitants to migrate into the urban centers. This migration is a major cause of the violent slums for which Brazil is famous.

The Drylands Polyculture Project

(Back to top)
It was in this dramatic scenario that the Drylands Polyculture Project was born. It was observed that despite its apparently barren prospects, the Sertão is a rich and prosperous land, producing many local and adapted crops and fodder plants. The problem was not the climate-as is usually implied-but the agricultural model, based on slash-and-burn, or worse, European practices of cleancropping and tilling, totally inappropriate for local conditions. In the old days, the small fields were never left bare, being a rich and dense mish-mash of short-, middle- and long-term crops of varying heights (corn, beans, tree cotton, castor bean, pumpkins, native cucumbers, cowpeas, sesame, sunflowers, peanuts, cassava; some of these crops reached tree height). Thus the overall destructive effects of the slash-and-burn were limited to small areas, and compensated for by this mixture of crops.

The Polyculture Model is, in part, an adaptation of the old model which in itself is an imitation of the local ecosystem, including the factors of density, biodiversity, and dynamism. As in the local ecosystem, there is a variety of legume trees to help fix nitrogen, considerable organic mass when the rains come, and woody ground cover (from tree prunings) during drought, protecting the land long after other organic material has decomposed. This guarantees permanent root masses which capture any drop of rain, and leaf surface to capture the dew that is an important source of humidity in this climate. These systems have demonstrated beyond a doubt that where there is vegetation there is water, and not the contrary-the denser the systems have been planted, the more humid they have remained, long into the dry season.

Using this local model of biodiversity, density, and dynamism, various combinations were implemented which contained the following elements: drought-hardy forage plants which guarantee some production even in El Niño years (such as Opuntia cactus); leguminous trees, some native, some adapted, such as Gliricidia and Leucaena; plants to fix nitrogen and produce biomass in general (jack beans (Canavalia) and pigeon pea (Cajanus cajan), and the above-mentioned leguminous trees); short-term catch crops (radishes, sesame) planted in furrows with a bit of compost or manure, and legume trees to replace them; castor bean (the main cash crop, resistant to drought); rows of corn; and cowpeas filling in the rest of the space. Ideally , there needs to be one plant every ten centimeters, especially in poor soils where the plants do not fill out well and/or during the first year. Cassava can also be included in this scheme, replacing one of the corn rows. Cane grass could be an important element because it is an excellent mulch grass when properly managed, but has not been included because the farmers fear its potential invasiveness and do not want to plant it.

thumbnail of planting diagram
Thumbnail link to planting diagram and picture of typical field, ~8K file

The fields are planted with cactus just before the rains come, four meters between rows, one meter between plants. This effectively marks the field (which can become quite complex!) so that the farmer can orient the rest of the crops properly. Normally, in the same row as the cactus, castor bean is planted in the dry soil, when the first signs of approaching rain appear (native cactus flowering, etc.) When the rains come, seeds of fruits, legumes, and fine hardwoods are planted together with the cactus, for the future agroforest, and sesame is interplanted with the castor bean, since the geometry of the two plants (castor bean round, sesame tall and straight) combines well and efficiently uses the space. The four meters between the castor bean rows are densely planted with food and legume crops, interspersed with leguminous trees that will shade and protect the space in the dry season.

thumbnail image of polyculture field, northeastern Brazil
Thumbnail link to picture of young polyculture field, ~48K file

The second year, when the rains are due again, the process is repeated, this time pruning the leguminous trees and any other plant (castor bean, pigeon pea) which survived the dry season and using the prunings as mulch to protect the soil. As a result, the soil is gradually built up, improving in fertility over the years.

Thus, from the same field, year after year, the farmer will harvest some vegetables, corn, beans and cowpeas, sesame, pigeon pea, and (as the trees mature) fruits for his table, while having castor bean for sale and fodder for his animals (taking care not to rob the field of its biomass! Any fodder harvested must be returned in the form of manure). In the worst years he may harvest mostly fodder plants, but this will guarantee that he will always have meat and milk to eat and income from the sale of the animals to buy other food, thus avoiding total collapse of the system. We believe that with this system even in the worst years he should also have some castor bean crop, giving enough income to get through the bad times and not have to migrate.

A second phase of this project will be the conservation of food for the dry season, especially hay for the animals and dried leaf meal for the humans. (This dried meal, produced from leaves of crops such as cowpeas, pigeon peas and cassava, is added to such foods as bean dishes and soups. Besides being tasty, it helps counter vitamin A deficiencies, which are common on the region.) Of course the family will also need ample cisterns to guarantee drinking water for the family and animals, an aspect which is being intensely addressed by local NGOs.

These polycultures are so healthy that absolutely no form of artificially produced chemical is necessary to maintain them. They have proved to be resistant to insects and other diseases, even when growing side-by-side with infected fields. The biomass and presence of legumes eliminate need for manures, which become an element of luxury, used, when available, to increase the production of the short-term crops.

thumbnail link to castor bean chart
Thumbnail link to chart of castor bean contribution to Brazilian GNP, ~9K file

This polyculture model, as developed by Ernst Gotsch, a Swiss farmer/researcher who has been developing agroforest models for Brazil for the last 15 years, is more than ten years old. However, the project of large-scale planting of demonstration fields only began this year. The project is financed by the local castor bean industry, which was interested in having a steadier supply of raw materials. These first demonstration fields have given spectacular results: one farmer harvested the equivalent of 1,400 Reais (about 700 dollars) from one hectare alone. This is about double the regional average income per hectare and compares very favorably to the state's minimum salary of 150 Reais per month or 1,800 Reais per year. Furthermore, this sum does not include the armfuls of corn and pumpkins that the farmer gave to his many curious visitors, and the food he harvested for his own table during this first season. He is so happy with the results that he plans to plant 10 hectares in the model next season. Thirty of his neighbors are also planning to try out the model. Not all the demonstration plots gave such spectacular yields, sometimes due to the severely degraded soils, sometimes due to the farmer not implementing the model as designed (they find it hard to believe that they can plant so densely and still harvest something!) But ALL plots produced significantly better than the conventional plots, and all promise to weather over this first dry season with some vegetative cover. This means that the soil is being protected and improved, promising better results for the coming year.

The success of these field runs has caught the attention of the state agriculture department and the local extension service, and even of one local bank, which is considering changing policy to demand the polyculture model as a condition for financing. When the next rains come, there are plans to plant 300 demonstration plots, with the direct involvement of the extension agents who then will be responsible for spreading the model even further.

With this simple yet powerful technique, we hope, in the space of a few years, to substantially lessen the tide of land degradation and mass migration from the Brazilian drylands, thus helping to rescue this rich area from social and economic collapse.

bar denoting end of article text

Author information

(Back to top)
Marsha Hanzi is the founder of the Bahian Permaculture Institute. She works with a multi-disciplinary team in helping farmers and new communities develop sustainable systems. In existence since 1992, the Institute has been responsible for training numerous professionals in agroforest techniques and sustainable design. They offer traineeships to students from around the world. The Institute will also host the Second Latinamerican Permaculture Congress at the end September 2001 in the Chapada Diamantina Mountains of Bahia.

Those interested can get in touch with her by email at
Web site (in Portuguese):

Additional web resources:

(Back to top)
Special: Biodiversity for Food and Agriculture
From the FAO's Sustainable Development database, this online document is extracted from "Human Nature: Agricultural Biodiversity and Farm-based Food Security" by Hope Shand, an independent study prepared by the Rural Advancement Foundation International (RAFI) for the Food and Agriculture Organization of the United Nations (December 1997).

Cultivating Diversity: Agrobiodiversity and Food Security
This 1998 report, prepared by the World Resources Institute and available from the United Nations Development Programme web site, summarizes the importance of agrobiodiversity to food security, productivity, and ecological sustainability. It proposes guidelines to help meet the worldwide need for food while safeguarding opportunities for farmers and their communities. NB: this is a large document!

About the Arid Lands Newsletter

Link to ALN home page Link to index page for back web issues Link to index page for pre-web issue archive Link to this issue's table of contents