Hacker News Reader

Comment by V__ | Hacker News Reader

> Solar deployment is now running at about $500 billion per year, which means that about 0.5% of global GDP is being spent on solar deployment. This figure is up an improbable 43% Y/Y

What a mind boggling relative and absolute increase.

View on HN

No, not mind-boggling at all. It's about a tenth of what is required to meet governments' signed-up-for targets for limiting global average temperature rise to 1.5 degrees above pre-industrial.

For reference about 2% of global GDP is spent on new fossil fuel mining and drilling investment, and another 2-4% on the fuels (including refining).

A 10x increase is about 6 years at the current growth rate. That's mind-boggling.

It can also be not enough. They're not mutually exclusive.

I would have expected the ratio of $ spent on new fossil fuels vs. new solar to be a lot higher than 4. My mind is quite boggled.

The figure I like to quote is that we need 2 to 5% of GDP spend per year to get to net zero by 2050.

We already pretty much passed 1.5C anyway, right?

Yes. If we were spending 5% of global GDP on Solar and Wind now, there'd be a slim chance. Scaling up to that in six years, as a sibling comment claims, is way too late.

This is caused by the cost of solar declining 89% in the last decade [1]. Batteries are next [2]. Which leads us to solar and batteries powering the world [3]. Extrapolate exponential growth of generation and storage, not linear. There is >1TW of generation and hundreds of GWs of renewables and storage (respectively) in aggregate across all US grid operator queues, for example [4]. Similar story in China [5].

(obvs we have a long way to go, just need to push the pedal to the floor; enough sunlight falls on the Earth in 2 minutes to power humanity for a year [6], and space is not a concern [7] [8])

[1] https://ourworldindata.org/grapher/solar-pv-prices

[2] https://ourworldindata.org/battery-price-decline

[3] https://news.ycombinator.com/item?id=37435387 ("Solar and batteries are going to win, and our thinking needs to adjust")

[4] https://www.pv-tech.org/nearly-1tw-of-renewables-in-us-inter... ("Nearly 1TW of renewables in US interconnection queues as wait times continue to grow)

[5] https://www.theguardian.com/world/2023/jun/29/china-wind-sol... ("China is set to double its capacity and produce 1,200 gigawatts of energy through wind and solar power by 2025, reaching its 2030 goal five years ahead of time; ...as of the first quarter of the year, China’s utility-scale solar capacity has reached 228GW, more than that of the rest of the world combined.")

[6] https://www.ku.ac.ae/two-minutes-of-sun-enough-to-power-a-ye...

[7] https://landartgenerator.org/blagi/archives/77565 ("Land and Ocean Areas to Support a 100% Renewable Energy, Zero-Emissions, Regenerative Global Economy")

[8] https://news.ycombinator.com/item?id=37492062 ("Declining populations free up ag land for solar in densely populated countries")

Holy cow that lithium price graph is astounding! It's a log plot that looks like a linear decline in price. It's like Moore's law for lithium..

Moore’s Law is a special case of the more general “Law of Accelerating Returns.”

> In my view, it is one manifestation (among many) of the exponential growth of the evolutionary process that is technology.

https://www.thekurzweillibrary.com/the-law-of-accelerating-r...

Great charts. I hope they update them for current years soon. I'm not sure if: the trend continued (because of technologic progress and the learning curve), the trend reversed (because of supply chain issues), or stayed the same because of the combined effects of those 2.

Very informative. Note from your 4th link I wasn't previously aware what "interconnection queues" were. This is a good overview: https://emp.lbl.gov/queues.

Importantly, though, only a small percentage of that capacity "in queue" will actually get built. From your link:

> Much of this proposed capacity will ultimately not be built, however, with only 23% of projects seeking connection from 2000 to 2016 having subsequently been built based on a LBNL analysis of a subset of queues. Only ISO-NE and ERCOT exceeded 30% completion rates, with CAISO performing the worst at 13%.

Good callout wrt queue completion ratios. Unfortunately, there aren't many comparable leading indicators of future generation to rely on (maybe EIA's 860M survey? [1] [2], but it only gives you a short look into the future). Plans are plans until steel is in the ground and glass is getting racked. Open to other suggestions if it improves modeling and forecasting.

Regardless, the amount of renewables and batteries coming online (at least in the US) can't be overstated [3].

[1] https://www.eia.gov/electricity/data/eia860m/

[2] https://www.eia.gov/electricity/monthly/

[3] https://www.eia.gov/electricity/monthly/images/figure_6_01_c...

> Regardless, the amount of renewables and batteries coming online (at least in the US) can't be overstated [3].

> [3] https://www.eia.gov/electricity/monthly/images/figure_6_01_c...

The "Other" generating unit type on that map is all over CA and TX. Looking at the source data for the map [1], it's mostly batteries. I'm surprised they didn't break those out as their own category.

1. https://www.eia.gov/electricity/data/eia860m/

Jevon's paradox is one hell of a thing. When you suddenly make something very cheap, a lot of people suddenly start wanting a lot of it.

What you are seeing is basically the consequence of solar changing from not the cheapest power source available into the cheapest power source available. Immediately, everybody that would invest into something else changed into solar.

Reply on news.ycombinator.com