Environmental impacts of organic and conventional agricultural products – Are the differences captured by life cycle assessment?. (1st February 2015)

• Record Type: Journal Article
• Authors: Meier, Matthias S.; Stoessel, Franziska; Jungbluth, Niels; Juraske, Ronnie; Schader, Christian; Stolze, Matthias
• Is Part Of: Journal of environmental management. Volume 149(2015:Feb.)
• Page Start: 193
• Page End: 208
• Publication Date: 2015-02-01
• Subjects: Comparative life cycle assessment (LCA) -- Organic -- Conventional -- Agriculture -- Attributional -- Consequential -- N-surplus; Environmental policy -- Periodicals; Environmental management -- Periodicals; Environment -- Periodicals; Ecology -- Periodicals; 363.705
• DOI: 10.1016/j.jenvman.2014.10.006
• Languages: English
• ISSNs: 0301-4797

Abstract

Abstract: Comprehensive assessment tools are needed that reliably describe environmental impacts of different agricultural systems in order to develop sustainable high yielding agricultural production systems with minimal impacts on the environment. Today, Life Cycle Assessment (LCA) is increasingly used to assess and compare the environmental sustainability of agricultural products from conventional and organic agriculture. However, LCA studies comparing agricultural products from conventional and organic farming systems report a wide variation in the resource efficiency of products from these systems. The studies show that impacts per area farmed land are usually less in organic systems, but related to the quantity produced impacts are often higher. We reviewed 34 comparative LCA studies of organic and conventional agricultural products to analyze whether this result is solely due to the usually lower yields in organic systems or also due to inaccurate modeling within LCA. Comparative LCAs on agricultural products from organic and conventional farming systems often do not adequately differentiate the specific characteristics of the respective farming system in the goal and scope definition and in the inventory analysis. Further, often only a limited number of impact categories are assessed within the impact assessment not allowing for a comprehensive environmental assessment. The most critical points we identified relate to the nitrogen (N) fluxes influencing acidification, eutrophication, and global warming potential, and biodiversity. Usually, N-emissions in LCA inventories of agricultural products are based on model calculations. Modeled N-emissions often do not correspond with the actual amount of N left in the system that may result in potential emissions. Reasons for this may be that N-models are not well adapted to the mode of action of organic fertilizers and that N-emission models often are built on assumptions from conventional agriculture leading to even greater deviances for organic systems between the amount of N calculated by emission models and the actual amount of N available for emissions. Improvements are needed regarding a more precise differentiation between farming systems and regarding the development of N emission models that better represent actual N-fluxes within different systems. We recommend adjusting N- and C-emissions during farmyard manure management and farmyard manure fertilization in plant production to the feed ration provided in the animal production of the respective farming system leading to different N- and C-compositions within the excrement. In the future, more representative background data on organic farming systems (e.g. N content of farmyard manure) should be generated and compiled so as to be available for use within LCA inventories. Finally, we recommend conducting consequential LCA – if possible – when using LCA for policy-making or strategic environmental planning to account for different functions of the analyzed farming systems. Highlights: We revealed considerable bias in comparative LCA studies on agricultural products. We suggest how to improve sustainability assessment of agricultural products in LCA. A more precise differentiation of farming systems is needed within LCA. In inventories we found large deviations between modeled and actual N-fluxes.