The project contributed to the important global question of how to reduce environmental impacts from food production, with a focus on South African food value chains. The Swiss and South African project team joined forces to ensure basic data collection for South African maize, dairy and fruit production, life cycle inventory generation and subsequently environmental hotspot analyses. The main results gained were the following: 1. Electricity consumption used for irrigation in maize production causes higher greenhouse gas emissions than dryland maize, due to the high percentage of coal-generated, fossil electricity. Using PV electricity for irrigation reduces the carbon footprint of the maize by 33%, making it comparable to the carbon footprint of dryland maize. 2. Direct emissions from enteric fermentation of dairy cows make up about 70% of the greenhouse gas emissions of South African raw milk. For other environmental impacts, such as non-renewable energy demand and eutrophication, the production of concentrated feed and the associated land use are more important. Using feed additives as methane inhibitors potentially reduces the carbon footprint by 18%. Differences between breeds were assessed for Holstein and Ayrshire cows, but due to inconsistent results, no sustainability recommendation could be developed thereto. 3. Electricity consumption for irrigation in fruit production and for the cold storage rooms and packhouses make up the highest share of greenhouse gas emissions, acidification, eutrophication and ecotoxicity. Accordingly, renewable energy use, from e.g. PV electricity, can effectively reduce the environmental impacts from South African fruits. In 2016-2017, we presented results at several conferences. Prof. Harro von Blottnitz organised a workshop in Stellenbosch, South Africa, on January 26th of 2017 where the research team from UCT and ZHAW presented project results to South African stakeholders. In the following period, two scientific publications were submitted to the open access journal Sustainability.