Are cows eating the Amazon?

Land clearing in tropical countries for production of export crops gets a lot of attention, and rightly so. However, the understanding of the mechanisms involved and how to allocate the effects of deforestation in terms of environmental damage or carbon emissions, is still very low. While it is true that exports are important for this, most deforestation are driven by domestic factors.  A study by Henders et al (2015) show that in Argentina, Bolivia, Brazil, Paraguay, Indonesia, Malaysia, and Papua New Guinea one third of deforestation was embodied in exports in 2011, up from a fifth in 2000. This means that two thirds of the deforestation is driven by domestic factors.

In the study beef was identified as the main driver of forest loss in the seven countries, accounting for nearly 60 percent of embodied deforestation and just over half of embodied emissions. Soybean production was the second largest source of embodied deforestation area whereas oil palm was the second largest source of embodied emissions.

But one can argue against how to allocate emissions and land use changes. For instance, pasture areas in Brazil have been stable in the last decade, while grazing has moved towards the forested areas because it is more profitable to plow and farm crops in the pasturelands than to graze them. It might be more correct to allocate impacts of land use changes based on the relative expansion of agriculture area for different crops or production in a country. In the case of Brazil, it would mean that soy production would still be a major cause of deforestation, but sugar cane cultivation would carry 20 % of the burden of deforestation and associated carbon emissions while beef from pasture would carry almost none of this, This would be the case despite that there is very little sugar grown in the recently deforested zones and a there are a lot of animals grazing there.

A study of de Ruiter et al (2016) goes a step further and state that “all global LUC [Land Use Change] emissions should be allocated to agricultural land itself, not only to recently cleared land, resulting in a calculated average emission of LUC for every hectare in agricultural use.” This means that an average hectare in Europe would have the same emission factor as a recently cleared patch in the Amazon – LUC emissions of 1.18 tonne of CO2e for every hectare of agricultural land.

One can of course argue back and forth; it is simply not the case that we can say that one way of counting is the correct one. But it can hardly be correct to let any “meat” carry the burden of deforestation in the Amazon; it is not primarily the consumption of Amazonian pastured beef, nor the fact that European pigs eat Brazilian soy beans that drive Brazilian deforestation. It is simply cheaper to import Brazilian soy than to feed European animals with European protein feed. Europe has taken much bigger areas of farm land and pasture out of production than Brazil has opened. It is still correct to allocate the emissions caused by deforestation to the agricultural system, but rather to the globalized trade than to the consumption of any particular commodity. One can make a similar argument about soy bean oil from Brazil or palm oil from Indonesia or Malaysia.

Another complication with all these calculations is that they are based on a rather static view of developments, while the reality is a lot more dynamic. For example in our farm we are now clearing 10 hectares of secondary forest for making pasture for cattle. If we allocate carbon emissions from deforestation per kg meat for a ten year period in a similar way as is mostly done in LCA analysis our meat will be responsible for several hundred kilograms of carbon dioxide equivalents per kg! But the wood will, perhaps, replace other wood in the paper pulp industry, or will be used for renewable energy. And regardless of that, the land was open and grazed some fifty years ago, even earlier it was a wetland (emitting a lot of methane). Sweden has lost more than a million hectares of pasture land in hundred years, and previously cattle, sheep and goat also grazed the forest which was not at al as dense as it is now. None of these aspects are covered well with the current approaches, neither with LCA analysis of a particular production, nor with macro-level studies as the ones referred to above. We actually get government (EU) support for restoring this pasture.

de Ruiter H, Macdiarmid JI, Matthews RB, Kastner T, Smith P. 2016 Global cropland and greenhouse gas impacts of UK food supply are increasingly located overseas. J. R. Soc.  Interface 13: 20151001.

Henders S. et al 2015, Trading forests: land-use change and carbon emissions embodied in production and exports of forest-risk commodities, Environ. Res. Lett. 10 (2015)