4 ms·
In the 1970s, the US had plans to build 1,000 nuclear power plants. If they were the average size, I believe would come to double our current electrical generat
by lkbm 5y ago
In the 1970s, the US had plans to build 1,000 nuclear power plants. If they were the average size, I believe would come to double our current electrical generation capacity. Where would we be with cheap, plentiful, carbon-neutral electricity?
"Electricity and heat" has been the dominant source of emissions in the US since at least 1990[0]. Even if most of that is the "heat" part, we can heat much more efficiency with electricity than with gas nearly the entire year in all our climates, and in the rare cases where that's not true, resistance heating can cover the offset. The only reason we use gas is because it was cheaper.
Second on the list is transportation, much of which is finally hitting viability for electrification right now. Again, if we'd had cheap, plentiful electricity, that would have spurred on development and implementation much earlier. Maybe not the 1980, but a decade or two sooner definitely seems believable.
There are likely some industrial processes that we can't easily switch over, but I think given a cost incentive, we'd find a way for most things.
I am kind of assuming the switchover would mean "cheap" electricity. Maybe we'd have plentiful expensive electricity, but I'd expect market forces to push the price down.
[0] https://ourworldindata.org/emissions-by-sector#annual-greenhouse-gas-emissions-by-sector https://ourworldindata.org/emissions-by-sector#annual-greenh...
- akamaka 5y agoHeating and electric vehicles are in fact the perfect example of the phenomenon I’m describing. Already today, it would be more efficient to heat your home with a heat pump powered by a natural gas generation station, than to burn that same gas for heat directly. Already today, it’s more efficient to convert fossil fuel into electricity and charge an electric car, than to burn that fuel in the engine. There are high barriers of technology, scalability, and economics to overcome to make this transition happen, and if we had built a lot more nuclear in the 80s and 90s those barriers would still be nearly as high.