Bioethanol is promoted as a means of tackling climate change, diversifying energy sources and securing energy supply. However, there also concerns that their wider deployment could lead to unintended environmental consequences. Life cycle assessment (LCA) is a widely used methodology to assess the environmental performance of biofuels. However, its outcomes strongly depend on the inventory data and modeling assumptions. Agronomic variables such as crop yields, nitrogen fertilizer rates or field emissions of nitrous oxide are very sensitive inputs, as are soil carbon dynamics in response to land use changes (LUC) entailed by the deployment of energy crops. Models simulating agroecosystem processes and the economics of agricultural farms are promising tools to predict such variables and improve the reliability of LCA.Here, we combined the agro-ecosystem model CERES-EGC, the farm economic model AROPAj and the LCA approach to investigate the effect of local drivers on the environmental impacts of bioethanol from miscanthus and switchgrass over France.Overall, lignocellulosic bioethanol achieved GHG abatement targets in the 74 %–94 % range compared to gasoline, and complied with the 50 % minimum imposed by European regulations. Miscanthus-based ethanol achieved up to twice lower environmental impacts than switchgrass due to 50 % higher biomass yields overall. Low fertilizer N input rates (in the 0-30 kg N ha-1 yr-1 range) proved the most efficient strategy to optimize energy return. Significant inter-regional variability occurred, especially in terms of soil C sequestration rates, which weighed in substantially on GHG budgets. Some regions were more efficient than others as a result, which advocates a site-specific approach and a potential prioritization when planning biorefineries, taking into account local production and environmental performance potentials. Compared to previous studies, ours provided high-resolution data in terms of crop yields, nitrous oxide emissions and soil C dynamics, factoring in LUC effects at local to regional scales.
Read full abstract