Reading the article, I thought that it was not relevant to compare carbon footprint per unit of mass, and I was pleased to see it was addressed in the last section. However, the way it's addressed feels even more disingenuous: they removed all the non protein-rich food from the chart. Naturally, mostly animal products remain at the top.

Based on the assumption that humans need protein[0], it seems only fair to compare everything we eat on a carbon-cost / protein-value basis without cherry-picking the data points. It's here:

https://ourworldindata.org/grapher/ghg-per-protein-poore?tab...

A few observations: - coffee and chocolate are very high up. To be fair, nobody assumes that those have any kind of nutritional value, and that's not why they are consumed. - beef and lamb are very high. It implies that, even in the most charitable light, lamb and beef are bad in terms of carbon emission[1]. - among non-animal products, tomatoes, cassava and bananas are surprisingly high up. Higher than a lot of animal products. Now that is relevant, because those are consumed for their actual nutritional value. Rice, a staple all around the world, is worse than farmed fish, or eggs! - wheat and nuts have an amazing carbon-cost/nutritional value ratio.

[0] and fat, there are no essentials carbs. [1] and that makes me sad because I love both.