Soybean production expansion scenarios

 I anticipate the conclusion: In the case of Brazil, there is no need to deforest to expand soy production. All we need to do is use and properly combine two alternatives that we have, which are the use of areas that have already been anthropized and the constant increase in crop productivity.

In the first case, the anthropized areas come mainly from pastures, whether degraded or released due to improvements in zootechnical indexes. It is estimated that, currently, there are more than 20 Mha of degraded pastures and, in the next 30 years, an area of ​​more than 30 million hectares of pastures, in good condition, will be released for other uses, much of it suitable for use in annual crops, such as soybeans and corn. Additionally, the increasing use of crop-livestock integration systems (iLP) also makes it possible to expand the soybean area, without the need to resort to deforestation.

How much will we produce?

On pg. 163 of the book “The soy saga – from 1050 BC to 2050 AD.”, published by Embrapa, the most likely scenario points to a doubling of global demand for soy between 2020 and 2050. It would be up to Brazil to play a leading role in this process, estimating that 260 Mt will be produced here in 2050. We can do some exercises to understand the importance of increasing productivity as a way of expanding soybean production without deforestation.

Initially, it is important to visit the CESB website and see how a growing number of producers have demonstrated that it is possible to go well beyond the Brazilian average for soybean productivity. The national record is 8.945 kg/ha, much higher than the highest Conab average, which was 3.394 kg/ha, in the 2016/17 harvest. And it's not just a matter of record, hundreds of producers from different regions have already demonstrated that it is possible to greatly increase soybean productivity, while increasing profitability. If these producers can do it – and they only use technologies available on the market – any other soy producer can do it.

How are we going to produce?

Table 1 lists several possibilities, from setting the current productivity (3.394 kg/ha) until 2050, to the possibility of not having any area expansion, with the increase in production being obtained exclusively through productivity gains. These two extreme hypotheses only serve to compose the curve, and attention must be paid to the central values ​​of annual productivity gains (between 1 and 1,5%).

In order to avoid area expansion, we would need average annual increases of more than 2% in productivity, reaching over 6.000 kg/ha in 2050. Although lower than the current productivity record, we consider that it is not feasible to obtain this average productivity, even in 2050. On the other hand, setting the highest average productivity ever obtained by farmers in Brazil (Conab), we would need to almost double the soybean area by 2050. This hypothesis can also be completely disregarded, as productivity will increase in the next 30 years.

In the book A Saga da Soja (pg. 39) it is verified that, between 1960 and 2018, there was an average geometric increase of 1,53% in soybean productivity. Therefore, based on the recent past, it can be considered feasible to obtain similar values ​​for the immediate future, considering the unusual advancement of Science in the coming decades. In this case, Brazilian soybean productivity in 2050 would be between 4.574 and 5.305 kg/ha, an increase of less than 50% over the highest average productivity ever obtained in Brazil. In this case, we would need to expand between 12 and 20 Mha by 2050. Based on the information on the existing stock of anthropized areas, and the potential to release 30 Mha of pastures by that date, we are very comfortable in stating that it is perfectly possible to produce the entire soybeans that Brazil demands by 2050, without cutting down a single tree.

And in the Cerrado?

I recently participated in two workshops promoted by The Nature Conservancy (TNC), whose objective was precisely to propose sustainable models for expanding soybean production in the Cerrado. This NGO projects an increase of 7 Mha for the Cerrado by 2030, in addition to the current 19 Mha.

The simulation shown in Table 2 shows that soybean production in the Cerrado is expected to expand from 64 to 88 Mt, between 2020 and 2030. Column C calculates the productivity required to obtain the same production (column B), without changing the area, which would require annual productivity gains of 3,2%. And, in column D, the productivity obtained with annual increments of 1,53% is presented, the same index obtained in Brazil between 1960 and 2018. In this case, the demand for area would be less than 50% of that predicted by the TNC, reaching , in 2050, productivity of 3.951 kg/ha, which is very likely and feasible.

The additional soybean area by 2030, going from 19 to 26 Mha (column A), can come exclusively from the release of pasture land, due to the constant gains in zootechnical indices. Thus, again, and by another line of reasoning, it is possible to conclude that Brazil can produce the soy that will potentially be demanded from the country, without cutting down a single tree. An integrated, persistent and permanent effort to gain sustainable productivity in soybean cultivation is urgently needed, based on research and development of technologies with this objective, and technical assistance to ensure that these technologies will be widely adopted.

By Décio Luiz Gazzoni, researcher at Embrapa Soja and member of the Sustainable Agro Scientific Council