3 ms·
> Hell, at that point the suburban areas could actually be net carbon sinks, given there's more room for big carbon-devouring trees in between single-family/dup
by notJim 7y ago
> Hell, at that point the suburban areas could actually be net carbon sinks, given there's more room for big carbon-devouring trees in between single-family/duplex/quadruplex homes than there typically is between skyscrapers packed as tightly as possible (as is the norm in even small cities, let alone large ones).
This is absurd. If people move away from the suburbs, the trees will still be there, and more if we get into reforestation. Building a suburb generally involves cutting down trees. Humans living in a smaller area of the planet leaves more room for trees.
- yellowapple 7y ago> Building a suburb generally involves cutting down trees. Only for cities/suburbs that exist in forested areas. This would exclude most desert and prairie ecosystems (for example). > Humans living in a smaller area of the planet leaves more room for trees. Sure, mathematically-speaking, and planet-wide. On a more local scale, though, this is not necessarily the case. There's also the consideration for relative density of CO₂ sources v. sinks. I don't have any actual data here, so feel free to ignore this, but my hypothesis is that urban areas would have significantly higher CO₂ densities than suburban areas (which would in turn have significantly higher densities than rural areas, and in turn wilderness areas), and this high-density CO₂ would not readily/easily/immediately disperse / would be more difficult for surrounding carbon-sink areas (like wilderness and - I'd argue - potentially suburban areas) to even receive, let alone absorb. You thus end up with big lingering blobs of CO₂, continuously fueled by human occupation, and unable to disperse. Spreading out humans would better spread that CO₂ sequestration load, thus allowing better efficiencies.