LCIA method choice shapes biodiversity footprints: comparing updated and established methods in a country-resolved assessment of German food consumption

Abstract Purpose Global food systems are a major driver of biodiversity loss, yet biodiversity footprint results can vary substantially with the life cycle impact assessment (LCIA) method applied. We quantify the potential biodiversity impacts of food production for German consumption in 2021 and test how conclusions on dominant drivers, product hotspots, and supplying regions depend on the LCIA method, distinguishing reliable (“no-regret”) from method-sensitive mitigation levers. Methods We integrate a trade-linked, product- and origin-specific life cycle inventory for German food consumption from the FABIO model (63 food products) with different LCIA methods. Keeping the inventory constant, we compare results for multiple drivers (land use, water use, freshwater eutrophication and climate change) across: GLAM (spatially explicit; country level), LC‑IMPACT (spatially explicit; country level), and ReCiPe2016 (global-average). Agreement and divergence are assessed using convergence indicators aligned with core footprinting purposes (main drivers, product hotspots, and geographic origins). Results Land use and water use dominate throughout all methods and water-use hotspot patterns are comparatively stable: product and country shares are similar between the spatially explicit methods, though absolute impacts differ markedly. Driver level analysis shows a substantial reweighting of land use impacts in GLAM (52%) vs. LC-IMPACT (15%) and ReCiPe2016 (82%). Land-use results are more method-sensitive, with substantial divergence in product- and origin-level attributions that shifts hotspot rankings and perceived responsibilities (GLAM: more distributed around Latin America vs. concentrated North America in LC-IMPACT). All agree on dominance of animal products but GLAM broadens the picture mainly elevating the (land use) relevance of tropical commodities (Coffee, Cocoa, Palmoil). Conclusions Biodiversity footprint conclusions for prioritisation are partly method-dependent: broad mitigation directions can be robust across LCIA methods (animal products, nuts and products), whereas finer product rankings (e.g. cocoa, coffee, palmoil) and region- or driver-specific priorities remain contingent on the chosen impact model. Footprint studies intended to inform strategy should therefore separate reliable, (“no regret”) levers from method-sensitive, second-tier priorities and explicitly communicate uncertainty in hotspot rankings arising from LCIA method choice.