Knowledge for better food systems

Increasing beef production could lower greenhouse gas emissions in Brazil if decoupled from deforestation

This very useful paper focuses on Brazilian beef production in the Cerrado grasslands region of Brazil makes an important contribution to the on-going debate about the merits of different livestock production systems, and of different consumption patterns.

It models what happens to beef related emissions in the region under a range of scenarios that hinge upon two main variables:

  1. Changes in demand for beef leading to corresponding increases or decreases in cattle production.  It looks at reductions of 10%, 20% and 30% (with poultry meat substituting for the beef) as well as increases of 10%, 20% and 30% both in relation to a reference projection. (NB: as a personal comment from the lead author, Rafael de Silva, for increases in overall output for DBAU+10%, DBAU+20% and DBAI+30% this translates into a 31%, 43% and 55% increase in beef production by 2030 relative to production in 2016). 
  2. Changes in deforestation policy: one scenario where deforestation is controlled/halted (DLD), and one where deforestation rates respond to changes in demand (CLD) – i.e. deforestation rates increase if demand increases, and decrease if demand falls off.

The study takes into account, in quantifying cattle-related emissions the following sources:  emissions or sequestration arising from positive or negative changes in soil organic carbon (SOC), deforestation-related emissions, direct livestock emissions (e.g. methane); emissions from use of agricultural machinery, fertilisers, feed production etc.

The study finds the following:

a. In a scenario where deforestation is controlled (i.e. deforestation is not allowed to happen in response to increases in demand for beef), then increased demand for beef actually leads to reductions in greenhouse gas emissions.  By contrast, if deforestation is controlled AND demand is reduced, then emissions actually increase. 

Why is this?

The study finds the following reason: if more beef is wanted, but the area allocated to beef production is not increased (because of controls on deforestation) then this means that production has to intensify.  Intensification is not achieved by feeding dedicated feed-grains to livestock since Brazilian beef are almost entirely grassfed, but rather through improving grazing management.  Using evidence from other research the study finds that intensifying grazing (in a well managed way) can stimulates the productivity of the Cerrado’s deep rooted grass, Brachiaria which in turn take up more carbon from the atmosphere.  And so the direct methane emissions from the cattle are more than compensated for by the net carbon uptake from the soil.  However, under a reduced demand scenario, while no deforestation occurs, there is not the same incentive to increase the productivity of the livestock system, meaning that the same benefit of carbon uptake from the grass doesn’t occur – indeed the grasslands become net emitters of carbon dioxide. To quote: “Lower demand and smaller herds require less grass production, reducing the incentive to maintain or increase productivity; pastures then degrade, losing organic matter and soil carbon stocks. Higher

demand combined with effective deforestation control policies leads to more efficient systems with lower emissions intensity due to significant increases in carbon uptake by deep-rooted grasses in improved pastures.”  Note that the authors assume that in the demand reduction scenario, people eat poultry meat instead of beef.  (NB. following an email exchange with Rafael, the lead author, it seems that emissions arising from broiler substitution account for 2% of the total emissions under DBAU-10% and 7% of total emissions under DBAU-30% (scenario with most substitution).   

b. In the scenario where deforestation is not controlled: then, predictably, the increased demand scenario looks considerably worse than the reduced demand scenario.  The study also says that while in the increased demand scenario deforestation emissions increase in proportion to the increase in demand, under a reduced deforestation scenario, the same is not so.  Deforestation rates can never be less than zero – i.e. the study doesn’t assume that afforestation might occur since currently there is no economic motivation for farmers to do this.

The study authors suggest however that scenario a is more likely, since the signs are that Brazil is achieving success in halting its deforestation, “a trend attributed to a combination of supply and demand side factors including intensification in large scale commodity-oriented farming, market regulation (for example, moratoria on beef and soy grown in recently opened areas), product certification, and more effective law enforcement.”

The authors also undertake a sensitivity analysis to look at what might happen if you get a combination of only partially-effective deforestation controls and only partial-intensification and identify a critical point of equilibrium, such that the increases in deforestation emissions are balanced by sequestration in the Cerrado grasslands.

The authors consider all this over a 150 year time scale to take account of the fact that soils reach equilibrium after a while – they can no longer take up more carbon.  They calculate that by about 2057, the Cerrado soils is approaching saturation point, after which the direct annual emissions from livestock will be greater than the sequestration from the grasslands.  They assume that demand increases will level off by about 2030.  However it is not until 2120 that accumulated emissions in the increased consumption scenario are greater than the accumulated emissions in the reduced consumption scenario.

Finally the authors conclude that “restoration of degraded pastures is the biggest opportunity for national mitigation plans; indeed, after avoided deforestation, the restoration of 15 Mha nationwide from 2010 to 2020 is the main measure contributing to the 40% reduction target by 2020.”   


Recent debate about agricultural greenhouse gas emissions mitigation highlights trade-offs inherent in the way we produce and consume food, with increasing scrutiny on emissions intensive livestock products. Although most research has focused on mitigation through improved productivity systemic interactions resulting from reduced beef production at the regional level are still unexplored. A detailed optimization model of beef production encompassing pasture degradation and recovery processes, animal and deforestation emissions, soil organic carbon (SOC) dynamics and upstream life-cycle inventory was developed and parameterized for the Brazilian Cerrado. Economic return was maximized considering two alternative scenarios: decoupled livestock–deforestation (DLD), assuming baseline deforestation rates controlled by effective policy; and coupled livestock–deforestation (CLD), where shifting beef demand alters deforestation rates. In DLD, reduced consumption actually leads to less productive beef systems, associated with higher emissions intensities and total emissions, whereas increased production leads to more efficient systems with boosted SOC stocks, reducing both per kilogram and total emissions. Under CLD, increased production leads to 60% higher emissions than in DLD. The results indicate the extent to which deforestation control contributes to sustainable intensification in Cerrado beef systems, and how alternative life-cycle analytical approaches result in significantly different emission estimates.


De Oliveira Silva, R., Barioni, L. G., Hall, A. J., Folegatti Matsuura, M., Zanett Albertinii, T., Fernandes, F. A., and Moran, D (2016). Increasing beef production could lower greenhouse gas emissions in Brazil if decoupled from deforestation, Nature Climate Change, 10.1038/NCLIMATE2916

We’d welcome your thoughts on this paper (when signed in, use the “add comment” box) which is worth reading in conjunction with Schader et al 2016 and with Elin Röös’ paper which she recently blogged about on the FCRN website.  They all discuss very similar issues but at different scales/in different contexts, and underpinned by somewhat different assumptions.

We also encourage you to read a related blog-post by one of the authors for the FCRN: Where’s the beef? Re-examining the nexus of livestock, deforestation and greenhouse gas emissions in Brazil.

You can find related resources in the Research Library categories on primary production: agricultureanimal issuesmeat, as well as the keywords categories on animal feedsustainable healthy dietsconsumption and production trendsgrazing and land use.

You can read related research by browsing the following categories of our research library:


mathros's picture
Submitted by mathros on

From Lobato et al. (2014):

Brazil beef cattle production systems can be considered as “grass-fed based”, since all breeding and rearing are made on pastures, and only 7.5% (Brazilian Association of Feedlot Producers), or even less, of the slaughtered cattle are finished on feedlots, and for a short period of time.  

I have checked the most recent data on that (Anualpec, 2015) and it is still estimated as less than 10% of finished cattle.

Lobato, J.F.P., Freitas, A.K., Devincenzi, T., Cardoso, L.L., Tarouco, J.U., Vieira, R.M., Dillenburg, D.R., Castro, I., 2014. Brazilian beef produced on pastures: sustainable and healthy. Meat Sci. 98, 336–45. doi:10.1016/j.meatsci.2014.06.022

Thanks for your comment 


Xavier Poux's picture

The only thing that is missing in the paper is the use of Cerrado under its native vegetation, i.e. savanna.

Indeed, a "sustainable intensification" based on bracharia and improved pastures might have a positive impact on soil, when compared to badly managed increase of stocking rate on native savanna.

However, it is unclear why reducing the meat production is necerally going along a poor land management: one can decrease the "improved" area and save native vegetation. In addition, extensive cattle management is considered as poor management, and Cerrado "wasteland", but it needs to be better assessed and probably not by the upholders of intensive management.

But the main points do not stand on these issues, these are:

- from a biodiversity conservation, native cerrado is one of the richest biome in the world. Its conservation is hampered by the case of Amazonia, but it is a crucial issue to conserve it. 

- from a C perspective, it is unclear what is the better land use. Native cerrado has the capacity to store C in the root systems of its vegetation;

- last but not least: increased biomass production, being through grass or planted trees (such as eucalyptus), has a major impact on the water balance. Native Cerrado is a system which does not grow much during the dry season and thus preserve the water balance. Alternative more productive and intensive uses, producing in the dry season, will unavoidably affect the water resource. Which is a major concern as the Cerrado is the main water reservoir of the whole Brazil and when anticipating climate change impacts.

- This quantitative aspect is combined with qualitative issues (i.e. use of pesticides in the whole crop rotation soya/corn/grass).

Generally speaking, savannas are more difficult to capture than forests in the C and biodiversity agenda. They appear as wasteland from an semi-intensive agricultural production (while they are extensive farmland or rangeland); they might appear as less efficient in terms from a C storage perspective. This "in between" position does not serve their cause.

But having made a recent field trip in the Cerrado near Brasilia, indeed beef production looks (african) grass fed.

mathros's picture
Submitted by mathros on

That is a quite a sensible observation of why lower demand would imply de-intensification in the scenario where beef is decoupled of deforestation.

The reason for that is pastures would reforest only if they were taken out of production entirely. Despite that could potentially be an ideal scenario, from our understanding, that is not likely to happen for the following reasons:

1) Farmers aim to maximise profit. Assuming demand is lower than predicted, the way to minimise costs is reducing investments to improve pastures as feeding cattle responds for ~80% of all beef production costs in a farm. Unless recovering natural vegetation would make higher economic return, but that is not the case.

2) Landowners typically maintaining larger than necessary pastures for speculative reasons when cattle ranching is a means to secure land ownership, but  aiming to sell  when the cropland frontier advances (Bowman et al., 2012; Strassburg et al., 2014)

3) Currently there are legal requirements enforcing a minimum stocking rate, with a penalty of expropriation if land is considered unproductive. Expropriated land is used for land reform with cattle ranching being a relatively simple land-use occupation (Bowman et al., 2012)

4) If natural vegetation is recovered there is a ban on re-clearing reforested areas (Strassburg et al., 2014) and farmers will take that into account when considering risk of reforestation.



Latin America and the Caribbean

Latin America and the Caribbean occupies the central and southern portion of the Americas. The region is home to the world’s largest river (the Amazon River), the largest rainforest (the Amazon Rainforest), and the longest mountain range (the Andes). Export-oriented agriculture constitutes an important part of the economy, especially in Brazil and Argentina. This large continent has a range of climates spanning the ice of Patagonia, the tropical forests of much of the continent, and more temperate regions in, for example, Mexico and Chile. Due to the greatly differing geography and economic development in the continent, all types of agriculture can be found in Latin America. Subsistence farming and cash cropping with coffee, cocoa and so on are common in many nations including most of central America, whereas large-scale beef production in the cerrado of Brazil provides an example of hyper-large farms run by large businesses.

View articles relating to South America


Doc Type