Life cycle-based environmental methods and indicators for sustainable consumption and production in New Zealand food systems : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Science in Life Cycle Management at Massey University, Manawatū, New Zealand

Abstract

The Sustainable Consumption and Production (SCP) approach requires considering the entire life cycle of a product – also known as life cycle thinking (LCT). Environmental Life Cycle Assessment (LCA) is one of the most widely used, internationally recognised, and standardised LCT-based SCP tools. The aim of this research therefore was to investigate how LCA can be used to support sustainable food systems, by applying it to both the production and consumption of food within a national context, in this case, New Zealand.

For the production-focused aspects of agri-food systems, the research focused on Voluntary Sustainability Standards (VSS) – a set of voluntary, predetermined standards with indicator-based criteria for use by agri-food producers. Most agri-food VSS criteria use practice-based rather than performance-based indicators, and they are mostly checkbox-type ones. Therefore, the first research question was, ‘What is the role of LCA in the development of performance-based indicators for VSS criteria for a New Zealand agri-food sector?’ The New Zealand avocado sector was used as a case study to address this question. A qualitative study was undertaken to explore the perceptions of stakeholders about the potential for a life cycle-based VSS for the sector. Most respondents identified the need to quantify the environmental impacts of the avocado sector and supported use of a life cycle perspective. They noted the weak representation of environmental sustainability in the standard criteria of Global G.A.P., the only certification scheme used in the sector. Thus, in summary, they supported the introduction of a complementary scheme with science-based performance metrics and including the whole value chain.

An extensive LCA study was then undertaken for the orchard stage of the New Zealand avocado value chain, to guide the development of indicators for facilitating continuous environmental improvement. Data were collected from 49 orchards across the three main growing regions in New Zealand, plus 4 orchards that were recently established and thus classified as being in the non-productive and low production years of the avocado orchard life cycle. Individual orchard, regional and national level impact scores were calculated to use as benchmarks in indicator development, the latter two enabling growers to see their relative ranking in environmental performance at regional and national level. The impact results varied widely between the orchards across the three regions, and further research is needed to develop a better understanding of the reasons for this variability. The study also modelled the other stages of the orchards’ life cycle, including the non-productive years of orchard creation, and the low-productive years during orchard establishment and senescence. The results from this model showed that the impacts increased by 13 to 26% per kg of avocados when including the low- and non-productive years of the orchard lifespan. Therefore, separate benchmarks using area-based functional units should be used for young orchards.

The system boundaries of the orchard-level study were then extended to include the post-harvest transport and packaging stage, as well as distribution to the first points of sale (PoS) in selected domestic markets and export markets in Australia and South Korea. The orchard stage was the hotspot for all the scenarios with the exception of some of the airfreighting scenarios. The contribution of the post-harvest stages to total impact values across the scenarios was nevertheless noteworthy, ranging from an average of 41% and 22% of total supply chain impacts up to the overseas port for export avocados and up to the regional distribution centre (RDC) for domestic supply respectively. Thus, it is appropriate to include the post-harvest stages when developing LCA-based criteria for a VSS for the New Zealand avocado sector. Second, the study also showed that the packaging types used for international and domestic distribution have quite different environmental impacts. Therefore, in a VSS context, having two separate benchmarks for these two would be a more realistic approach than using a weighted average. Third, sensitivity analysis results showed that the average airfreighting impacts across different export scenarios for each impact category were 31-924% higher than the alternative shipping values. Thus, with respect to future marketing strategies of the New Zealand avocado sector (of which the main trend is to increasingly diversify into Asian markets), an indicator for a future VSS will be influenced less by market destination, and more by the mode of transport used. It may be more useful to develop distribution-related benchmarks per unit of fruit transported over a set distance (for example, for climate change, kg CO₂ eq. per t.km) instead of using a distance-based weighted average per kg avocados. This could drive transport-related improvements like switching to slow shipping and electric vehicles, and optimising routes to reduce resource consumption and emissions. Lastly, sensitivity analysis also showed that extending the system boundary beyond the export port to retailers in Australia increased impacts, ranging from a 4% increase in water scarcity up to a 59% increase in terrestrial ecotoxicity. However, improvements in these downstream stages can only be instigated and monitored via a VSS if the stakeholders at the different stages represented in the VSS are cooperating together on this initiative. The New Zealand avocado sector stakeholders have more influence on the supply chain stages up to the export destination, but beyond this stage other stakeholders are more influential in terms of delivering environmental performance improvement with respect to these downstream activities. Therefore, it would be more feasible to maintain the system boundary to the port in terms of indicator development for a VSS for the avocado sector in New Zealand.

For the consumption-focused aspects of agri-food systems, nutritional LCA (nLCA) is a relatively new approach for conducting LCA studies where nutrition is considered the main, or one of the main functions of food. Thus, the second research question was, ‘What is the role of nLCA in helping New Zealand consumers make informed consumption decisions based on the combined nutritional value and environmental impacts of food?’ A study was conducted to investigate the development and potential application of an nLCA method to rank meals, using a case study of a ‘toppings on toast’ (ToTs) meal consumed in New Zealand. Nutritional value was calculated using the NRF family of indices – specifically the NRF9.3 and NRF28.3 indices. A quartile-based method was used to assign weights to the NRF and climate change scores, which enabled ranking of the toppings. It was found that the rankings of the ToTs based on the nLCA results did not change much between the two indices. This suggests that the NRF9.3 index may be sufficient to determine overall the best, medium, and worst performing toppings in the ToT meal context. However, the results also showed that water-soluble vitamins and unsaturated fats included in the NRF28.3 index contributed notably to the nutritional scores for most of the toppings and were instrumental in the rank changes for the toppings which are particularly rich in these nutrients. Thus, for a more diverse range of toppings/meals, an expanded index including these nutrients can generate more nuanced rankings.

General insights on VSS indicator development include the importance of including post-harvest stages; of conducting individual LCA studies for different agri-food sectors; of using non LCIA-based performance metrics for certain aspects like ecotoxicity assessment and airfreighting; and applying spatially disaggregated characterisation factors for water use. At the orchard stage, the importance of considering input and impact variability as well as the early and late years of the orchard were important factors for indicator development. At the post-harvest stage, the need for separate indicators and benchmarks for export and domestic packaging was noted, as well as the possibility of using impacts per t.km as an indicator for non-airfreight transport. At the consumption end, the need for updated nutrition intake data in New Zealand was highlighted. A nutritional index with fewer nutrients (but including mono and polyunsaturated fats) could be appropriate for practical application of the nLCA method in New Zealand. Further, nLCA results can be presented to consumers, based on both serve size and energy content, ideally via a graded label based on single scores or using a quantile-based colour-coded table if required by the consumer.

In summary, the research identified key considerations for use of LCA-based indicators and benchmarks in a VSS scheme to support the New Zealand avocado sector and linked these to the wider implications for agri-food production systems. For more sustainable food consumption, a new environmental rating methodology for alternative meals was developed based on nLCA, and its usefulness to consumers demonstrated using a case study of a toppings-on-toast meal in New Zealand. Thus, this thesis contributes to the agri-food SCP literature with respect to both method development and its application in New Zealand.