In recent years, scientists have repeatedly concluded that meat, dairy, and egg production are worse for the environment than plant-based food production. However, a detailed look at how those differences look across different products, environmental indicators, and in different world regions has been missing so far.

Soy milk might be better for the environment than cows’ milk, but to what extent exactly? And is switching from cows’ milk to soy milk better for both avoiding climate change and using less water at the same time? Are there differences in how eco-friendly a milk is depending on where it is produced (e.g., in North America vs in Europe)? And how do different production methods compare?

In a new research paper published in Science, researchers compiled numerous results from previous studies and life cycle assessments. To new levels of detail, this paper examines the environmental impacts of different food products in different countries. One of the major take-aways is as follows: even the least-impactful animal products are worse for the environment than the average plant-based product.

The authors compiled data from 38,700 farms in 119 countries. They also covered product value chains from 1,600 processors, packaging types, and retailers. They compared the environmental impacts of 40 food products, which represents close to 90% of the global consumption of protein and calories. The analysis of the environmental impacts included five indicators: greenhouse gas emissions (GHG), land use, acidification, eutrophication, and water use. The abundance of data allowed for fine-grained analyses.

Let’s take, for example, the GHG emissions from producing one liter of cows’ milk compared to one liter of soy milk. The average difference is already large: for soy milk, the GHG emissions are less than a third of those of cows’ milk (1.0 vs 3.2 kg CO₂eq per liter). But the picture becomes even more striking if we look beyond averages. While the GHG emissions from most (5^th to 95^th percentile) soy milk production are between 0.6 and 1.7 kg CO₂eq per liter, the range for cows’ milk is much wider—between 1.7 and 7.0 kg CO₂eq per liter. The comparison looks even worse for cows’ milk if we look at the other four environmental indicators.

After analyzing the data for 40 products across seven food groups—protein-rich products (including meat and legumes), milks, starch-rich products, oils, vegetables, fruits, sugars, alcoholic beverages, and stimulants—the authors drew some general conclusions. With regard to animal agriculture, they noted multiple reasons why plant-based protein is better for the environment than animal protein, and why these reasons will likely stay into the future:

• Emissions from feed production exceed emissions from plant-based protein farming.
• Two-thirds of deforestation for agriculture is done for growing feed for farmed animals. Deforestation releases previously captured carbon, which fuels climate change.
• Animal-specific emissions come from animals’ digestion processes, manure, and aquaculture ponds.
• The emissions from transporting and killing animals, especially from slaughterhouses, are greater than the emissions from processing most non-animal products.
• During production, more fresh animal products are wasted than other products.

The authors also discussed how different changes in production methods could reduce environmental impacts. While there is some wiggle room in terms of how humans could produce food from animals in a less environmentally damaging way, this approach has its limits. The authors note that changes to what we eat are crucial:

Moving from current diets to a diet that excludes animal products has transformative potential, reducing food’s land use by 3.1 billion hectares (a 76% reduction), including a 19% reduction in arable land; food’s GHG emissions by 6.6 billion metric tons of CO₂eq (a 49% reduction).

Also, global emissions relating to acidification and eutrophication would be cut by half, and fresh-water withdrawals by one fifth. In the US, the potential gains are even greater:

For the United States, where per capita meat consumption is three times the global average, dietary change has the potential for a far greater effect on food’s different emissions, reducing them by 61-73%.

This study builds on several others published in recent years, all pointing to the same conclusion: if we really care for the environment, we need to change the way we produce food and move toward diets that don’t involve the rearing and killing of animals. Animal advocates have one more comprehensive analysis to emphasize this point with. Also, advocates can draw on the publicly available data from this study to tailor their message to regional or product-specific needs.