[ben_w]

PV can supply in the order of one thousand times current worldwide all-source power consumption, even when the panels are limited to 20% efficiency.

Storage and transmission still have room for improvement, but generation by PV is already so cheap that all we have to do to “win” that part of the race is continue the last 30 years of trends in PV deployment for roughly 12 more years.

[skyfaller]

We definitely need to move to solar power. That said, there are some serious issues with the transition. One of them is that constructing and deploying solar panels inevitably produces greenhouse gases, and there is a limit to how much solar power we can produce per year while remaining within our carbon budget (since we are starting the transition so late): https://www.lowtechmagazine.com/2015/04/how-sustainable-is-p...

As a result, switching to renewables needs to be combined with reducing our energy use, so that renewables can meet our needs without causing the emissions that we are seeking to eliminate on a very short timescale. Reducing energy use, as you might imagine, is a very complex subject.

One difficulty is the Jevons Paradox: https://en.wikipedia.org/wiki/Jevons_paradox If you rely too heavily on the technological solution of energy efficiency, people just use the technology more and energy usage isn't actually reduced. Low-Tech Mag mentions the example of more efficient LED lights leading to the popularization of digital billboards, which use more energy than a regular billboard (but less energy than a digital billboard would have used before LEDs). https://www.lowtechmagazine.com/2018/01/bedazzled-by-energy-...

[ben_w]

Your first link is complaining that we are collectively building them in the wrong place (CO2 powered) and installing them in the wrong place (suboptimal sun).

As China is now the leading producer and consumer of PV, I suggest this is self-correcting, and that their PV production will naturally become greener — the more units they make and install the greener each subsequent unit becomes to make.

[skyfaller]

Regardless of whether PV production is in fact self-correcting (that seems like an empirical question requiring data), what I took away from "How Sustainable is PV solar power?" is that people aren't aware there is a problem, let alone the scale of it. I think people assume we can just add solar panels until we have enough solar panels to cover our current energy needs, but:

- energy demand is increasing

- solar panel production + installation creates emissions, and may have been a net negative at some times/places in the past

- we can't spend our entire carbon budget on solar panels

- yet we must dramatically increase renewable energy very quickly to avoid climate catastrophe

Personally I'm hopeful that technology like Heliogen's solar oven will help solar panel manufacturers eliminate fossil fuels from production, by achieving very high temperatures sustainably: https://www.cnn.com/2019/11/19/business/heliogen-solar-energ...

But I fear that if there is not enough awareness of or investment in the issue, we will blow past our carbon budget due to not carefully considering exactly how we execute the transition to renewable energy.

[ben_w]

Right now I would say that all of those points are true but also misleading.

The first unit of PV that you build in an all-coal economy will necessarily be a very CO2-intensive manufacturing process; a unit of PV made in a PV-only (electric mining equipment also) economy will have none.

The power output of that first particular unit can supply enough energy over its lifetime to build between 10 and 60 more of the same unit depending on which estimates I use for lifetime and energy cost per unit.

China is heavily dominated by coal, I think just under 2% PV by actual generation? So yes, very dirty. But their growth over the last decade was about x670, more than enough to turn the dirty production green in the next decade. (Far too much even for that, which is why I think their rates of increase have gone down in the last few years).