Land use & biodiversity

Evidence quality

Evidence that livestock occupies most agricultural land globally is high because it is supported by large, harmonized datasets spanning many products and production systems (Poore and Nemecek, 2018). Evidence on future land sparing and biodiversity recovery is model-based and therefore conditional on dietary, policy, trade, yield, and restoration assumptions (Leclère et al., 2020; Kozicka et al., 2023).

Global land use

Poore and Nemecek synthesized data from approximately 38,000 farms in 119 countries and evaluated land use alongside greenhouse-gas emissions, acidification, eutrophication, and water use. Their global analysis estimated that meat, aquaculture, eggs, and dairy use about 83% of farmland while providing 37% of protein and 18% of calories. It also found wide variation among producers of the same product, but even lower-impact animal products often exceeded vegetable substitutes across several environmental indicators (Poore and Nemecek, 2018).

Livestock land demand includes both grazing land and cropland used to produce feed. Consequently, comparing only the cropland used for foods directly consumed by people omits pasture and feed production and does not represent total agricultural land use (Poore and Nemecek, 2018).

Biodiversity pathways

Land-use change affects biodiversity primarily through habitat conversion and changes in land management. An ensemble of land-use and biodiversity models found that expanded conservation and restoration could reverse average modeled terrestrial biodiversity trends caused by habitat conversion, but the strongest outcomes required simultaneous changes across production, consumption, and food waste (Leclère et al., 2020).

Livestock can also affect biodiversity within land that remains under grazing. A global meta-analysis of forest livestock disturbance reported adverse average effects on forest structure, functions, and biodiversity, while a separate multi-trophic meta-analysis found that grazing responses varied with intensity, taxonomic group, and environmental context. These findings support the inherited caveat that extensive systems can have different interactions from intensive or poorly managed systems; they do not support treating every grazing system as ecologically identical (Li and Jiang, 2021; Filazzola et al., 2020).

Dietary change and restoration potential

A global economic land-use model examined replacement of pork, chicken, beef, and milk with nutritionally designed alternatives. In the modeled 50% substitution scenario, net loss of forests and other natural land was almost halted by 2050, and restoration of spared land within forest ecosystems could contribute to biodiversity and climate goals. The model also found large regional differences and stated that land-use policies would be needed to realize restoration rather than assuming that abandoned agricultural land automatically becomes habitat (Kozicka et al., 2023).

The distinction between land sparing and biodiversity recovery is therefore material. Lower agricultural demand creates the physical possibility of protecting or restoring land, while actual recovery depends on governance, ecological condition, time lags, and the restoration actions applied (Leclère et al., 2020; Kozicka et al., 2023).