Are healthy diets bad for the environment?

Have you seen the headlines recently?

• Vegetarian and 'healthy' diets could be more harmful to the environment, researchers say (Science Daily)
• Lettuce is ‘three times worse than bacon' for emissions and vegetarian diets could be bad for environment (Independent)

Could this really be?

The main difference between the current and healthy diets was in the amount of sugar and fats consumed.
These reports are based on research from Carnegie Mellon university published in the journal Environment Sciences and Decisions [1]. I found that paper shocking but not for the reason in the headlines. The paper actually evaluates the impact of switching to a more 'healthy' diet that is not particularly vegetarian. In fact the 'healthy' diet has more calories from animal products rather than less. But the main difference between the current and healthy diets is in the amount of sugar and fats consumed. The healthy diet has 400 less calories/day from 'added sugars' (equivalent to 25 spoons) and another 400 less calories/day from fats and oils. What I find most shocking is that people are eating this in the first place.

Replacing bacon with lettuce would give you vitamin A poisoning.
By the way, I can believe that lettuce might be higher in carbon emissions than bacon, but this is completely irrelevant. If you wanted to cut down on bacon you would be ill advised to replace this with lettuce. To get the same calories as 100g bacon you would need to eat 3.6 kg lettuce and you would have just consumed 53 times the recommended daily allowance of vitamin A. You could die of liver failure.



The current American diet includes 34% wastage and has double the recommended limit for sugar and fat.
The total calories count for the American diet was 2390 calories but the researchers added on another 34% for wastage giving a total of 3620 calories/day. Therefore a reduction of 400 calories of sugar is about 11%. Presumable this still leaves some sugar in the diet - the UK recommendation is no more than 11% sugars [2] so that means Americans are probably consuming double the recommended daily limit for sugars, and the same for fats.

The charts below are from the journal paper.

Change in calories consumption for the diets under evaluation [1]

Impact of the different food groups in terms of energy use, water footprint and GHG emissions [1]

Replace sugar and fat with dairy increases emissions, replacing them with grains and nuts would decrease them.
You can see that the recommended diet (scenario 3, with a healthier reduced calorie consumption as well as a better mix) involves 300 calories more from dairy products but these have 4 or 5 times the GHG emissions of the same calories from fat or sugar. This obviously increases the overall impact substantially. However if the empty sugar/fat calories were replaced with grains, or nuts and seeds, then the emissions would be reduced.

When replacing meat you need to worry about protein, not calories.
The study did not look at the impact of being vegetarian at all. Most people who go vegetarian replace meat with a mixture of beans, lentils and nuts/seeds. They may also eat some more dairy products. When replacing meat you need to worry about the amount of protein rather than calories. The Carnegie Mellon study did not separate out beans and lentils as a food group so I sought more information elsewhere - the chart below shows GHG emissions per gramme of protein based on data from 'The meat eaters guide' from the Environmental Working Group [3].

GHG emissions for different protein sources, per g protein.
Based on GHG emissions from [3] and protein content from Google.

Lentils and beans have 5-10 times less impact than beef.
You can see from this that for the same quantity of protein, lentils and beans have 5-10 times less impact than beef and 7 - 14 times less impact than lamb. Lentils are similar to chicken and turkey. Cheese and eggs are not as good as chicken but much better than beef or lamb.

Eating more healthily increases carbon emissions for this particular definition of 'healthy' eating.
Sadly this is not the first time I have seen a press release giving a completely misleading representation of a published paper. In this case, the paper was about healthy eating and nothing to do with switching to a vegetarian diet. The comparison between bacon and lettuce was headline grabbing but irrelevant. The conclusion of the paper, that eating more healthily would increase carbon emissions was true only for one particular definition of 'healthy' eating. Their introduction lists many other studies that show that more healthy diets, some vegetarian some not, tend to have lower carbon emissions.

Changing our food choices is one of the easiest ways to reduce carbon emissions.
I am not vegetarian but I eat about half the meat I used to and considerably less beef and lamb. I find a semi vegetarian diet much more varied and tasty, as well as reducing my environmental impact. There are loads of recipes around. For example tomorrow I will make sweet potatoes with red pepper sauce, possibly adding a few peanuts for protein. Changing our food choices has to be one of the easiest ways to reduce carbon emissions.


[1] Michelle S. Tom, Paul S. Fischbeck, Chris T. Hendrickson Energy use, blue water footprint, and greenhouse gas emissions for current food consumption patterns and dietary recommendations in the US (Environmental Sciences and Decisions) November 2015
[2] New National Diet and Nutrition Survey shows UK population is eating too much sugar, saturated fat and salt (www.gov.uk) May 2014
[3] Meat Eaters Guide: Climate and Environmental Impacts (Environmental Working Group)