Defining sustainable diets by comparing greenhouse gas emissions from different food groups: a systematic review

Abstract

BackgroundMany physical and biological systems are changing because of anthropogenic global warming, but many effects of climate change can be avoided, reduced, or delayed by mitigation. The UK Climate Change Act 2008 sets a target to cut the total annual greenhouse gas emissions (GHGE) by 80% by 2050, with an interim target of a reduction of 34% by 2020, compared with 1990 levels. UK food production and consumption is responsible for up to 18–30% of the total GHGE. Therefore, defining sustainable healthy diets is important to cut down GHGE from food. We have adequate information on nutrients in food, but we do not have robust data on GHGEs of individual food groups. MethodsThis systematic review estimated GHGE values (mean and range) for all the food groups in a commonly used food frequency questionnaire (FFQ). Eight medical and geographical databases were searched from 1995 to 2012 to identify lifecycle assessment (LCA) studies of food. Full papers or reports of selected studies were obtained. If they estimated the GHGE (individual gases or the total global warming potential [GWP]) per functional unit (eg, kilogram or tonne of meat, litre of milk) of any food group by LCA they were included. If studies assessed only emissions per land area or per farm they were excluded. We selected studies that described these components and authors were contacted if these components were not explained in the full report. Studies were not excluded if authors did not respond or were not available in English language. GWP per kg of food, in kg of carbon dioxide equivalence over 100 years (KgCO2e), was recorded from each study with details about farming methods, location, and the boundaries of lifecycle. The initial search identified 14 096 hits. 126 were selected after screening titles and abstracts and a further 42 papers were added by searching organisational websites and databases. Quality assessors were masked. FindingsBeef (mean 28·00, range 8·10–156·00) and lamb (15·52, 0·67–56·70) had the highest GWP values per kg and bread (–0·56, 0·25–0·93), fruits (–0·53, 0·06–1·37), and potatoes (–0·21, 0·13–0·48) were among the lowest values. The mean GWP for 1 kg of fruit juice was 2·92 KgCO2e. Whether organically and locally produced foods always have less GWP than non-organic foods and imported foods, respectively, is unclear. The GHGE values for organic products such as pork, chicken, eggs, potatoes, and tomatoes were higher than those for these foods when non-organically produced. The boundary of the LCA considered by the authors contributed substantially to variance in the results. Variation in results is due to a wide range of production systems, locations, and different system boundaries used in LCA studies. Therefore, the final estimates are not directly comparable and ranges overlap between food groups. GHGE is not the only indicator to measure the sustainability of food. However, GHGE was used in this study because it is the most commonly used indicator and government targets are set to measure the reduction in GHGE. InterpretationOrganic foods and locally produced foods do not always produce fewer GHGEs. Little consistency in LCA methods makes direct comparisons of estimates difficult. Future LCA food studies should make efforts to follow a uniform approach, to include common definition of stages in the lifecycle and inclusion of similar activities always under each category of food. These GHGE estimates are for common food groups in a FFQ and could be combined with existing nutrition databases to address questions around sustainable healthy diets. These findings will be used to quantify the GHGE changes of different dietary scenarios in the UK to achieve GHGE reduction targets. FundingBritish Heart Foundation.