Pathways for recent Cerrado soybean expansion: extending the soy moratorium and implementing integrated crop livestock systems with soybeans

Abstract

The Brazilian Soy Moratorium has effectively reduced forest conversion for soybeans in Amazonia.
\nThis has come at the expense of the region’s pasturelands, which have increasingly ceded space for
\ncompliant soy expansion. The question of extending the policy to the Cerrado, where recent soy
\nexpansion has come at the cost of ecologically valuable vegetation, plugs into a wider discussion on
\nhow to reconcile competing commodities on finite amounts of cleared area. Innovative management
\nstrategies that allow different land uses to coexist are urgently needed. Integrated crop-livestock
\nsystems with soybeans(ICLS)rotates beef and soy on the same area, and shows promise as a means to
\nimprove production, farmer benefit, and environmental impacts. Here we reconstruct historical land
\nuse maps to estimate Cerrado Soy Moratorium outcomes with benchmark years in 2008 and 2014, we
\nthen estimate additional production afforded by ICLS implementation between 2008 and 2014. We
\nfind that if a 2008 Cerrado Soy Moratorium were in place, 0.7 Mha of 2014 Cerrado soy area would
\ncurrently be in violation of the policy. Roughly 96% of this acreage is found in Matopiba (82%) and
\nMato Grosso (14%)states, suggesting that adoption may have slowed recent production in these
\nrapidly transforming soy centers, in contrast to central and southwestern Cerrado where there is more
\nconcentrated eligible expansion area. Changing the benchmark to 2014 could have added 0.7 Mha of
\neligible expansion area, though over 80% of these additions would be in states with the most 2008
\neligible area (Distrito Federal, Mato Grosso, Maranhão, Minas Gerais, Mato Grosso do Sul).
\nMeanwhile, ICLS adoption could have added between 4.0 and 32 Mha of new soy land to the study
\narea without additional clearing between 2008 and 2014, though this would depend on rigorous
\naccompanying land zoning policy to guide implementation. The roughly 5 Mha of Cerrado soybean
\nexpansion that actually occurred between 2008 and 2014 could have been accommodated on 2008
\nsuitable pasture area given an ICLS rotation frequency of every 6 years or less. Conservation estimates
\npresented here represent the upper limit of what is possible, as our scenario modeling does not account
\nfor variables such as leakage, laundering, or rebound effects.