> Surely, but they clearly states that cities pollute far more than large spread areas No, your source is highly specific in that it is looking at grid emissio

> Surely, but they clearly states that cities pollute far more than large spread areas No, your source is highly specific in that it is looking at grid emissions from cities in Australia, which has one of the dirtiest grids in the developed world at 527 gCO2/kWh. For comparison, the US is at 379, Canada is at 124, France is at 68, and the other European countries are somewhere in the 200-350 range. The results are very sensitive to this. The source even specifically covers some other things that make it irrelevant to extrapolate: > With regard to energy consumption overall warm weather causes may make Australia a special case due to its climate being very different to that in northern countries. If dwellings are too close together they are more difficult to cool whereas it is easier to heat them. The figure it uses to calculate emissions from rail transport is 105 gram per passenger-km. For Eurostar, that value is 6grams, due to higher ridership and much cleaner grid. Plus, the lowest carbon form of transport you can use is your own two feet. Next up is a bike, followed by an ebike. All of these rely on density, and are poorly represented due to sprawling development patterns in Aus (and North America).