A quick search gives me numbers around 10-15% of the total food energy. But if you remove processing, some of the packaging and some of the retail energy consum

A quick search gives me numbers around 10-15% of the total food energy. But if you remove processing, some of the packaging and some of the retail energy consumption, that are in large part consequences of not having access to fresh food, this number is much larger. The benefit of building smaller cities is that you can afford to have lettuce that comes from a couple kilometers away. If you build small cities, but still keep a distribution network that spans across continents, things will indeed not improve. Also you compare dense cities to (american?) suburbs. Dense cities are indeed better. My point is that they are a local optimum, and one that is not good enough to offset the effects of climate change. The model I would promote is cities small enough to rely on a short-distance farm network for most of their needs.