Land transformation and occupation impacts of farming practices for the production of soybean in Mato Grosso, Brazil, using life cycle impact assessment Michael J. Lathuillière1 ([email protected]
), Eduardo J. Miranda2, Eduardo G. Couto2, and Mark S. Johnson1,3 1Institute for Resources, Environment and Sustainability, University of British Columbia (Vancouver, BC Canada) 2Departamento de Solos, Universidade Federal de Mato Grosso (Cuiabá, MT Brazil) 3Department of Earth, Ocean and Atmospheric Sciences, University of British Columbia (Vancouver, BC Canada)
The production of soybean in the Brazilian state of Mato Grosso has been increasing steadily since 2000, expanding into natural ecosystems (tropical forest and savanna) as well as occupying land previously cleared for pasture. While land transformation for soybean production has been well studied and documented (e.g. Macedo et al 2012) little is known on the environmental and health impacts of production besides the detailed quantification of greenhouse gas emissions. This study assesses the cradle-to-farm gate potential impacts of soybean produced in the year 2010 by applying life cycle assessment (LCA) to derive a first estimate of impacts to human health and ecosystem quality from the multiple stages of soybean production in the state. Given the relative importance of supply chain interventions on the reduction of deforestation (Nepstad et al 2014), product LCA can highlight impact “hot spots” as key steps for improvements in the production process.
100% 90% 80% 70% 60% 50% 40% 30% 20%
Environmental impacts of soybean produced in 2010 were obtained using information collected from 110 farms in the state of Mato Grosso along with soil information (Shannguan et al 2014) and land transformation observations (Macedo et al 2012). This inventory data was then used in the IMPACT World+TM (v. 0.04) life cycle impact assessment model together with guidelines from Koellner et al (2013) for land transformation and occupation impacts (Fig. 1).
Life cycle impact assessment e.g. eutrophication, acidification, etc.
Fertilizer + seeds
Fig 2. a. Mato Grosso and its 2010 soybean production delimited per municipality. Values represent the number of farms whose data were used in this study; b. Impacts of soybean production in 2010 on human health (Disability Adjusted Life Years) and Ecosystem Quality (Potentially Disappeared Fraction of species m2 y) using IMPACT World+TM (v. 0.04) expressed per tonne of soybean. 100%
Goal and Scope definition Inventory Data e.g. fertilizer use, diesel consumption, etc.
Fig 1. a. General steps of a life cycle assessment (LCA) according to ISO 14044:2006. Data from the 110 farms in Mato Grosso represent the inventory data stage to which is applied the impact assessment model (IMPACT World+TM) to derive impacts of production per tonne of soybean; b. Framework for assessing impacts from land transformation (T) and occupation (O) (Koellner et al 2013). The soybean land use (S) ecosystem quality is compared to the natural vegetation (NV, as tropical forest or savanna) in order to derive the potential impact to biodiversity and ecosystem services. Land transformation and occupation impacts were allocated to the 2000–2010 soybean harvests keeping in mind land transitions described in Macedo et al (2012) and an average regeneration time of 56 years for tropical forest and savanna biomes.
Preliminary Conclusions The cultivation phase of soybean production dominates impacts on human health and ecosystem quality, followed by farm operations Fertilizer and pesticide production account for 12% of the total impacts of soybean production Occupation impacts were greater in the tropical forest despite greater soybean production in the savanna Tropical forest and savanna biomes share equal burdens of most transformation impacts, except for groundwater recharge and climate regulation for which tropical forests carried more weight to the overall impact due to changes in evapotranspiration and above- and below-ground biomass
0% 0% BDP Tropical Forest
-20% Tropical Forest
Fig. 3. a. Annual land occupation impacts per tonne of soybean produced in 2010 in tropical forest and savanna biomes; BDP= biodiversity damage (% Potentially Disappeared Fraction of species), ER = erosion (tonnes of soil potentially eroded); MF = loss in mechanical filtration capacity (cm d-1); PF = loss in physiochemical filtration capacity (cmol ha kg soil-1); GWR = increase in groundwater recharge (mm y-1); BPP = loss in biotic production potential (ton of soil C y); CRP = change in climate regulation (ton C equivalent lost to the air). b. Land transformation impacts for soybean produced in 2010 taking into account 2000–2010 land transformation reported by Macedo et al (2012) and subtracting allocations made to pasture. References Koellner T et al 2013 UNEP-SETAC guideline on global land use impact assessment on biodiversity and ecosystem services in LCA Int J of Life Cycle Assess 18 1188–1202 Macedo M N et al 2012 Decoupling of deforestation and soy production in the southern Amazon during the late 2000s Proc Natl Acad Sci USA 4 1341–1346 Nepstad D et al 2014 Slowing Amazon deforestation through public policy and interventions in beef and soy supply chains Science 344 6188–1123 Shannguan W et al (2014) A global soil data set for earth system modeling J Adv Model Earth Syst 6 249–263
Acknowledgments: work supported by the Belmont Forum and the G8 Research Councils Freshwater Security grant G9PJ-437376-2012 through the NSERC to MSJ. We thank the collaboration of the Associação dos Produtores de Soja e Milho de Mato Grosso (Aprosoja/MT) for kindly sharing information, as well as Cécile Bulle and Manuele Margni for helpful input.