I reply here since I see no reply under the right message, sorry > Source? This says otherwise, especially if you take into account you'll have to do less km i

I reply here since I see no reply under the right message, sorry > Source? This says otherwise, especially if you take into account you'll have to do less km in a dense city. If you go by bike it's even less. Personal considerations and experience, I think no such real study exists mostly because it should be really large. Taking into account all the variable in a real life case ALONE demand much effort. For first you should find data for a significant timeframe, let's say 10 years, measuring the raw materials needed to build a metro in some city, and I think no such data exists, you can get some projects but not a trustable bill of material. Secondly you have to track all buses and trains and their average consumption. Done this you have a generic measure of "how much it cost in capex and opex the public transport system of YouNameIt city", what you miss is a "YouNameIt city" spread equivalent with only private vehicles. In this case you can only measure a large number of people consumption in a spread area, count the cost of the original road network witch likely exists since so much time no one can count and their operational costs. Even if you reach such goal you have just a "first degree measure", you need to take into account resources needed to produce the resources you used to build the city and the spread area transport network, some are easy, let's say cement and sand, some others are far harder, like steel. And so on. You probably need an decade to craft a working model for a state like North Korea (a good one because due to it's politics (Juche) anything there is mostly "domestic", they do commerce of course but much less than other countries and they still have cities and spread are, though get real data on NK might be a bit hard). To cut it short the "carbon footprint" is a nonsense, sometimes useful but not an answer: a train travel it's far less energy intensive on a modern rail than a truck travel, the material travel, but thing change if you have to count resources needed to build and keep rails vs roads, including exchange nodes, because we do not have rails on the home front door or aside of a cultivated field. If you just count how many tons of steel you can start imaging what cost more and what less. At this scale we have no simulation, non real scientific study, only "I like it let's try" without possible comparisons. > You're comparing with a new build house, compare with a new/renovated apartment. I do on purpose because my new home was built in around two months, rebuild the building of my ancient apartment is practically impossible because not only you have to relocate ~53 families for an year or more but also because it's in the middle of a dense EU city where such large demolition than construction size would destroy the traffic of a significant part of the city for an year or more, such EU buildings was simply built considering an eternal useful life, and that's a bubble waiting to explode with damage not much different than those of a nuclear war. In USA something similar have happened with certain big condos collapsed or at least damaged enough to be beyond repair but USA cities tend to be far less dense and younger than EU one so MAYBE some might be rebuilt. The point is essentially that small buildings with enough ground around can be practically rebuilt, dense area can only be rebuilt in batch like relocating all the people from NY to another place and rebuilt the whole NY, witch I bet no one think it's doable or at least reasonable to do... > You still need them with spread, only more roads to spread everything around. Not exactly the same thing, yes, maritime commerce need harbors and having many little harbors and many little ships it's not much sustainable, so you need some "hubs" in the logistic network but having much more room for local production you can do them, and the rest of the logistic network is spread anyway so MUCH simpler and cheaper to be build and evolve. A simple example around me there are enough potential farm to have cows for meat, milk and cheese for the nearby residents, it's definitively not enough for modern living but in a hypothetical spread society it would be normal and actually it was normal in a not so far past where at least milk was generally local instead of import in from 1000-2000km away simply because there it's cheaper than here even with the needed logistic. Do so on scale and it's easy to view you need much less resources. We would anyway import shrimps from South America and India, but the volumes changes much.