Mineral resources depletion, dissipation and accessibility in LCA: a critical analysis

The depletion-based ADP method has been extensively used for more than 20 years in LCA. A discussion on the core concepts and assumptions it is grounded on is however still needed. Several methods were recently developed to address reduction of resources accessibility and dissipation in LCA; namely, EDP, ADR/LPST, ARP, and JRC-LCI, complemented by price-based JRC to capture value loss. This article aims at critically analyzing these methods.

It first describes the methods, in terms of concept, target, resource flows in scope, midpoint and/or endpoint impact mechanism, and temporal scope. It then performs an in-depth analysis of these methods, in order (i) to discuss assumptions and modelling choices from a user perspective (LCA practitioners and decision-makers) and (ii) to highlight and discuss the scientific-evidence and scenarios on which they build.

This article gives an overview of the methods’ relevance to the safeguard subject, model robustness, data quality and completeness, and operability. The ADP_{Ultimate Reserves}, EDP, and ARP methods rely on several assumptions, at the same time core in the modelling and scientifically questionable for some aspects. ADP_{Ultimate Reserves} and EDP methods both do not address the loss of value of mineral resources, while the JRC-LCI (and potentially ARP) methods combined with JRC price-based, and the ADR/LPST endpoint, do (considering economic value). Overall two schools of thought are identified, regarding recent methods that strive addressing mineral resources use in LCA: a first one building on currently available LCI databases, implementing assumptions and proxies (ADR/LPST, EDP, ARP), and a second one pledging for new LCIs to be developed (JRC-LCI method).

This article highlights and challenges some of the key underlying choices and assumptions on which the reviewed methods are based. Intentionally, it does not provide any recommendation on which methods shall be implemented. Yet we encourage authoritative initiatives in the LCA domain (e.g., UNEP GLAM or PEF) and LCA practitioners to cautiously and critically select LCIA methods. We moreover highlight that the approach undertaken by the “first school” is certainly easier to operationalize in the short-term, though it necessarily requires proxies to overcome unfit-for-purposes existing LCI datasets. On the contrary, the effort initiated by the “second school” calls for a deep change of paradigm on modelling mineral resources in LCIs, which certainly requires a longer timeframe for implementation. The latter however looks very promising for LCAs to be truly supportive of more resource-efficient products and systems.