To what extent are short food supply chains (SFSCs) environmentally friendly? Application to French apple distribution using Life Cycle Assessment

There is growing interest in re-localization and re-connection of agriculture and food consumption, and Short Food Supply Chains (SFSCs) are becoming more and more popular. However, there are few studies on their environmental performance. Existing studies focus primarily on comparing imports and domestic consumption, often according to a single environmental criterion (i.e., energy or carbon footprint), without considering the great diversity of subnational commercialization patterns. This paper aims at assessing the environmental sustainability of different archetypes of food supply chains, from global ones to short ones, to identify hotspots and discuss the conditions under which a given supply chain performs better than another one. The overall methodology is based on a full Life Cycle Assessment (LCA) with a focus on a fresh and unprocessed product: apples purchased in an urban area. First, a consistent definition and classification of supply chains, is provided based on geographical and organizational features. An innovative approach is then developed to compute logistics data representative of these supply chains, using Geographic Information System tools. Finally, a comparison of the environmental performances of archetypes of apple supply chains is provided. The results show the relatively good environmental performance of the national long food supply chain which is used as the reference scenario in this study. Moreover, there are great differences in the environmental performance of SFSCs. Direct off-farm sales have the same level of performance as the reference. On the other hand, direct on-farm sales can be very impactful. Results also highlight the impacts of the final consumer trip which are significant and highly variable, depending on consumer-retailer distance, weight of apples purchased, and transport means used. This variability leads to reconsidering the questions frequently asked in LCAs of systems with extreme sensitivity to highly variable parameters. The concern is no longer whether one scenario is better than another, but to determine the values of those parameters that allow for better performance. Focusing on these parameters has direct implications in terms of decision-making by providing straightforward results with operational recommendations that are understandable to the general public, and not only LCA indicators. (C) 2020 Elsevier Ltd. All rights reserved.