Publications

Food is wasted or lost in large quantities with significant economic, social and environmental implications. Food waste is present at every level of the agri-food chain. Environmental and agricultural policies must take into account clear measures and targets, as well as monitoring mechanisms to prevent and reduce food waste. The European project WASTEWISE, “Waste Avoidance Strategies for Environmental Sustainability”, is implemented by a consortium of 9 partners from 6 countries. The project is a multidisciplinary one, aligned with the current policies and objectives of the European Union on the environment and food waste, with a multi-stakeholder approach. One of the directions of the project is to provide sustainable solutions for food waste prevention and reduction.

Food waste prevention is increasingly framed as a sustainability solution, yet most life cycle assessments (LCAs) fail to account for rebound effects, i.e. systemic responses that offset intended environmental gains. While well-documented in energy field, rebound effects remain underexplored in food waste studies, despite their potential to substantially offset the climate benefits of interventions. This is especially relevant for interventions involving end-consumers, where monetary savings can be substantial. Previous studies show considerable rebound magnitudes. Salemdeeb et al. (2017) estimated that household re-spending could offset 60% of greenhouse gas savings. Meshulam et al. (2021) found that peer-to-peer food sharing triggered a 94% rebound, nearly cancelling the intended emission benefits. Sundin et al. (2023) found income-driven rebound offset 31–64% of climate benefits from surplus food redistribution. Hegwood et al. (2023) reported price-driven rebounds offsetting 53–71% of expected savings globally. Albizzati et al. (2022) showed that economy-wide feedbacks reduced climate savings by 38% through land use and trade shifts. These findings underline the risk of overstating environmental mitigation in policy evaluations.
This study synthesizes findings from a semi-systematic review, identifying five rebound types relevant to food waste prevention:
1. Income-driven rebound: Savings from waste prevention are re-spent on other goods/services.
2. Income-dependent rebound: Rebound effects vary across income groups.
3. Technological rebound: Waste reduction technologies increase energy/material use.
4. Price-driven rebound: Efficiency leads to lower food prices, increasing demand.
5. Economy-wide rebound: Systemic shifts in supply, demand, trade, or land use trigger additional impacts.
The absence of rebound accounting in food waste LCAs risks overstating the benefits of such measures and overlooking trade-offs between climate mitigation and social outcomes. We illustrate this with empirical examples and present preliminary insights from an ongoing survey among food system and food waste LCA practitioners. Initial responses show limited incorporation of rebound effects in practice, despite a general recognition of their
relevance. To address this gap, we propose a framework applying system expansion within attributional LCA to model income-driven rebound effects. By extending the system boundary to include environmental impacts from additional consumption enabled by cost savings, and using primary data (e.g. surveys), expenditure data, marginal budget shares, or scenario-based assumptions to estimate these effects, rebound can be modeled without relying on macroeconomic tools. Although this use of system expansion departs from its conventional purpose of resolving multi-functionality, it provides a practical and transparent means to capture behavioral consequences in attributional LCAs. This deviation is justified by the need to reflect real-world consumer responses to cost savings, which are otherwise excluded in standard attributional modelling but can substantially offset intended climate benefits. By integrating rebound effects into attributional LCA, our approach improves the credibility of ffood waste prevention assessments and supports more realistic climate policy design.

Food waste represents a significant environmental challenge, and therefore accurately quantifying the environmental impacts embedded in food waste is essential for evaluating and developing effective mitigation strategies. Currently, the EU member states are required to report the total quantities of food waste generated through different stages of supply chain, but the magnitude of environmental impacts embedded in the food waste are not quantified.
Linking food waste data with LCA data poses challenges due to differences in granularity, structure, and nomenclature. Here, we present a systematic methodology that integrates reported food waste data with established Life Cycle Assessment (LCA) databases to comprehensively evaluate the environmental burdens associated with food waste across supply chains. Our approach links food waste data with environmental impact data from LCA
databases using the FoodEx2 classification and hierarchy. This enables assessment at a detailed product level as well as aggregation into broader product groups through the FoodEx2 hierarchical structure. By establishing data linkage protocols and applying consistent impact assessment method, we enable assessment of spatially resolved estimates of several different environmental impacts embedded in wasted food in different supply chain stages in Europe. The same systematic approach can also be followed in environmental impact assessment of more detailed supply chain specific impacts, where the data is collected directly for example from consumers or industry. The integration of reported food waste quantities with LCA data enables stakeholders to identify priority intervention points and supports policy development aimed at reducing food waste and its environmental consequences. This systematic approach advances the state of food waste impact assessment by bridging data gaps and enhancing the comparability of environmental impact assessments.

Reducing the environmental impacts of the food system while providing adequate nutrition is a critical global challenge. Addressing this requires integrating environmental impacts from Life Cycle Assessment (LCA) databases with food composition and consumption data. This enables cross-disciplinary analyses that combine nutritional and environmental perspectives. The WASTEWISE project aims to provide robust environmental insights into food waste streams, a key component of the food system. A significant challenge encountered in the project is linking detailed food waste data, classified using the FoodEx2 standard, with the ecoinvent, Agribalyse and WFLDB databases which are sources for LCA background data, but which lack a consistent classification system. This discrepancy presents a substantial data matching issue.
Traditional approaches to bridging such gaps often rely on manual mapping, fuzzy matching, or semantic matching techniques as employed by Furrer et al. (2024). While somewhat effective, these methods are typically labor-intensive and require programming expertise. We explore a novel approach using Large Language Models (LLMs) to automate the linking process between FoodEx2-classified food waste data and the unstructured entries in the ecoinvent database. Specifically, we employed the Gemini 2.5 Pro model to match entries across the two datasets. Preliminary results suggest that the LLM-based approach performs well in accurately identifying corresponding entries,  highlighting its potential to streamline and enhance the efficiency of data integration in LCA studies. This LLM-based matching approach demonstrates broader applicability and can be a valuable tool for resolving differences in classification nomenclature and integrating diverse datasets in other scientific domains. The method offers a scalable alternative to conventional techniques, enabling more detailed and accurate environmental assessments that are accessible to a broader range of experts.