an interesting graphic

[totient]

From Lawrence Livermore National Labs, old link an interesting graphic showing the flow of energy in the US (units are, I believe, 10^15 BTUs).

It implies some unsurprising things, like we should stop driving SUVs and turn off the lights when we leave the room. But it also implies some more interesting things:

• US natural gas usage exceeds demand domestic supply; switching from oil to gas might improve efficiency some but isn't going to change our petroleum imports much.
  
• Electricity generation and distribution is nearly as inefficient as a passenger car and mostly coal-fired to boot. Plug-in hybrids don't help the big picture unless they're also part of some kind of distribution efficiency improvement.
  
• Freight is pretty efficient comparatively speaking. Dicking around with how we power semis is not going to help much.
  
• Doubling solar every 18 months (the current growth rate and not coincidentally equal to the doubling time in Moore's Law) won't even be noticeable for over a decade. It will take WWII-level increases in production for this to make a difference.
  
• Doubling wind could make a difference sooner than that, and even fractional differences in nuclear output would make a big difference.
  
• Geothermal makes a bigger difference than you'd think.
  

Comments

[xthread.livejournal.com]

Hrrm... I need to go dig up primary sources, I may be misremembering. To come out ahead, you just need for the efficiency of the electric vehicle, after taking into account the losses due to sucky distribution infrastructure, to significantly exceed the efficiency of the gas engine, presuming that the energy cost of the electric vehicle manufacture isn't dramatically worse or better than the energy cost of building gasoline vehicles. I would expect that highly concentrated pollution from coal electricity production would be easier to mitigate than broadly distributed pollution from internal combustion engines; also, we mostly can't turn off gas engines while we're stuck in traffic, but our electric vehicles mostly don't draw power when we're stopped, so there should be a bunch of efficiency-cost reduction hiding in there, even though calling it 'efficiency' is a bit of a misnomer.

[palmwiz.livejournal.com]

And my assertion here is that the efficiency of a perfect electric vehicle after taking into account the (shown in this chart to be terrible) losses due to sucky distribution infrastructure is not very much better than the efficiency of an average internal combustion vehicle, and perhaps worse than the efficiency of an above-average internal combustion vehicle. TCO of electric vehicles is so low because they're really running on coal and coal is cheap.

[xthread.livejournal.com]

But doesn't that chart say that the energy loss due to transmission et al is higher for petroleum than it is for generated electricity?

[palmwiz.livejournal.com]

The chart says that electricity transmission losses are a little more than 2/3, and light-duty vehicle losses are 80%, which makes the perfect plug-in hybrid a bit more than half again better than the average light-duty vehicle. But there is such a range of efficiency within light-duty vehicles, and such a preponderance of needlessly inefficient ones in the fleet, that I have a hard time believing that there aren't non-hybrid vehicles that manage to be as efficient as even the best vehicle operating in electric mode.

[xthread.livejournal.com]

Walk me through your math? I want to make sure I'm reading the chart the same way that you are..