If I'm reading the methodology correctly, the paper is mainly comparing the relative findings within each study. (They do have some other comparisons that don't, yes, but they are mainly looking at relative numbers where each is computed with the same methodology)

Our focus on the percent change from a diet switch relative to the environmental impacts of the baseline omnivorous diet described in each study, makes the findings comparable across papers. Within each paper, the environmental impacts of one diet are comparable to those of another diet because these are expressed as a function of calories provided, taking as a benchmark a requirement of between 2000 and 2700 kcal/person/day

They then look at the distribution of the relative change figures. The entire range looked at here is lower emissions

We can also look at non-review studies as well. Here's one comparing emissions of farming types more directly

The aim is to compare the environmental impacts of different diets with different levels of animal product consumption, while accounting for the type of farming systems (organic or conventional) of the food consumed.

A positive link between animal-sourced food consumption and total environmental impact was observed in this large sample of French adults. By far, omnivorous had the highest-level of greenhouse gas emissions, cumulative energy demand and land occupation while vegan diets had the lowest

We found that a 100% organic omnivorous diet exhibited higher environmental pressures, suggesting that following an organic diet without changing towards a more plant-based diet is of little help, at least as regards the studied indicators

the vegan diet, whatever the indicator considered, remained less resource-intensive and environmentally damaging than other diets

https://www.sciencedirect.com/science/article/abs/pii/S2352550919304920