For context, a back of the envelope calculation is:
* Solar energy (on earth) gives about 250 W/m^2 [0]
* Earth has an approximate radius of 6.371 * 10^6 m
* Estimating sunlight on a disk of earth's radius yields ~ 700 * 10^15 (Wh/day) (3.14159 * (6.371 * 10^6)^2 (m^2) * (240 W/m^2) * (24 h/day))
That is, the earth's budget is just under 1 exa (Wh/day).
Earth's population is 8.2 B people and under a very generous energy consumption of 30 (kWh/day), that gives approximately 250 (TWh/day) (8.2 * 10^9 (ppl) * 30 * 10^3 ~ 250 * 10^12 (kWh/day/ppl)).
In other words, we're using about 1/1000 of a (back-of-the-envelope) theoretical upper limit of solar energy available to us on a daily basis.
Only 70% of the incident sunlight enters the Earth’s energy budget—the rest immediately bounces off of clouds and atmosphere and land without being absorbed. Also, being land creatures, we might consider confining our solar panels to land, occupying 28% of the total globe. Finally, we note that solar photovoltaics and solar thermal plants tend to operate around 15% efficiency. Let’s assume 20% for this calculation. The net effect is about 7,000 TW, about 600 times our current use. Lots of headroom, yes?
When would we run into this limit at a 2.3% growth rate? Recall that we expand by a factor of ten every hundred years, so in 200 years, we operate at 100 times the current level, and we reach 7,000 TW in 275 years. 275 years may seem long on a single human timescale, but it really is not that long for a civilization. And think about the world we have just created: every square meter of land is covered in photovoltaic panels! Where do we grow food?
Seriously, if you haven't read his take on things yet, at least the first few posts are a must-read. It's on par with the Arithmetic, Population, and Energy lecture at UC Boulder by Al Bartlett (popularly titled "The Most Important Video You'll Ever See", which is less hyperbole than you might think; the lecture is riveting)[2].
To very TL;DR things: solar and tidal energy (and their derivatives like wind) are essentially the only sources of energy we can rely on as our energy requirements grow. We are shockingly close (~300 years) to measurably raising the equilibrium temperature of earth's surface through purely thermodynamic effects if energy use trends continue. This is completely independent of greenhouse gases, and assumes that Earth is a perfect blackbody radiator. Once we exhaust our energy budget from these sources, that's it. No magical unobtanium source of energy can solve the fact that producing additional energy on the surface of the Earth will raise its temperature. We will stop increasing our energy use one way or another once we hit this wall.
If we want to continue using more energy we'll need a whole second Earth to do it on. Great, we've colonized mars! What does that get us? Based on a 2.3% growth rate and the Rule of 70[3], we'll use up that second Earth in thirty years. We'll now need two Earths to keep growing for the next thirty years.
Assuming steady exponential curves over hundreds of years always results in nonsense. Going backwards you find out energy consumption doesn’t fit that model. For example the US total electricity consumption over the last 20 years should have gone up by 1-1.023^20 = 57.6% when it’s almost flat over that timeframe. Efficiency gains matter, as we don’t want to heat our homes to unlimited temperatures jump a comfortable one.
Globally rather than an exponential curve instead the global quality of life keeps rising as more people enjoy the benefits of modern technology like AC and tablets, but the number of people isn’t continually increasing. Birth rates keep declining so in the short term its populations catching up to increased lifespans.
stouset · 1h ago
It is not assuming that steady exponential curves will continue. The argument is that exponential growth cannot and so will not continue.
But then you run into other problems. Can an economy like ours—which is wholly predicated upon unbounded exponential growth—continue indefinitely when energy use is effectively capped? Yes, there are efficiency gains and productivity gains to be made. Yes, our population growth is slowing and fill eventually flatline or even decline. Those will extend the length of time before we fully exhaust Earth's energy budget. Growth will end, and likely on a significantly shorter timescale than recorded history.
> which is wholly predicated upon unbounded exponential growth
This is a false assumption. The economy is based on doing what people want. If people suddenly want glow in the dark kitchens someone will ramp up production of glow in the dark paint and take customers from companies that didn’t follow the trend. That’s the feedback mechanism keeping the economy functioning.
Growth at the micro level and growth at the macro level aren’t the same thing.
abetusk · 1h ago
The 250 W/m^2 is measured sunlight, not the amount reflected by the atmosphere.
I absolutely would not confine ourselves to land as oceans provide a large area available for solar energy capture and there's no reason to think we might not be able to use it.
Photovoltaics, or whatever technology we use to capture sunlight, will get better but even at a modest 20% still gives us a lot of headroom.
Whether its 2.3% or 2.5% energy growth per year, the calculation gives us a timeline of 200-400 years. For some reason this is used as a countdown to oblivion instead of a rallying cry about where we're headed. Mars is great but there's a *lot of space in space*. Besides pushing solar panels into orbit, either the earth, moon or sun directly, there's also tons of asteroids, ripe for mining.
Forget a second earth, we can make a Dyson swarm. What you take as a countdown timer to a bomb, I take as a timeline for us to go up the Kardashev scale.
These reduction to absurdity arguments about the temperature of the earth assume we're not going to space. I don't understand why you reject this idea outright.
All the above calculations about the energy available to us are from a tiny pin-prick sliver of how much energy the sun deposits in all directions, every day, all day, for the 4 billion years. Once we have access to a significant fraction of the suns energy, going to other stars and repeating is well within feasibility.
philipkglass · 1h ago
The "Galactic-Scale Energy" post is a great illustration that a constant percentage growth rate eventually hits hard limits imposed by physics.
Posts like abetusk's are a great illustration that "the solar budget" is a very generous energy budget. That may seem too obvious to mention, but in 20th century ecology literature (or even as recently as the early 2010s) living within "the solar budget" was often conflated with a low-energy, deindustrialized future. Constant growth fueled by sunlight (or anything else) can't go on indefinitely, but there's also no prospect that a sunlight-fueled world would have less energy available than the old fossil-fueled one.
stouset · 1h ago
It can't go on indefinitely, but thanks to exponential growth it also can't go on for much longer. 275 years is—to me at least—a shockingly short time-frame. Obviously it won't be something I live to see, but humanity being faced with insurmountable physical limits to growth in a handful of generations was eye-opening for me.
eunoia · 1h ago
> We will stop increasing our energy use one way or another once we hit this wall.
> If we want to continue using more energy we'll need a whole second Earth to do it on
Or we make like Niven's Puppeteers and move the Earth out into a further orbit with less insolation.
kulahan · 1h ago
Is there some reason nuclear isn’t considered while wind is?
zahlman · 27m ago
Wind energy ultimately results from solar energy: solar radiation differentially heats the atmosphere, causing atmospheric gases to expand, causing a pressure gradient, etc. Similarly, energy from our Sun enables plants and animals to grow, producing carbon-rich materials (fossil fuels, wood etc.) when they die that can be burned to release energy.
Nuclear power, on the other hand, releases energy stored in fissile heavy elements that were produced by the death of previous stars. There is no natural process using incident light from the Sun to create them.
kulahan · 12m ago
That makes complete sense - thank you for explaining. It's correct that nuclear would not be in this group here.
* Solar energy (on earth) gives about 250 W/m^2 [0]
* Earth has an approximate radius of 6.371 * 10^6 m
* Estimating sunlight on a disk of earth's radius yields ~ 700 * 10^15 (Wh/day) (3.14159 * (6.371 * 10^6)^2 (m^2) * (240 W/m^2) * (24 h/day))
That is, the earth's budget is just under 1 exa (Wh/day).
Earth's population is 8.2 B people and under a very generous energy consumption of 30 (kWh/day), that gives approximately 250 (TWh/day) (8.2 * 10^9 (ppl) * 30 * 10^3 ~ 250 * 10^12 (kWh/day/ppl)).
In other words, we're using about 1/1000 of a (back-of-the-envelope) theoretical upper limit of solar energy available to us on a daily basis.
[0] https://www.solar-electric.com/learning-center/solar-insolat...
To very TL;DR things: solar and tidal energy (and their derivatives like wind) are essentially the only sources of energy we can rely on as our energy requirements grow. We are shockingly close (~300 years) to measurably raising the equilibrium temperature of earth's surface through purely thermodynamic effects if energy use trends continue. This is completely independent of greenhouse gases, and assumes that Earth is a perfect blackbody radiator. Once we exhaust our energy budget from these sources, that's it. No magical unobtanium source of energy can solve the fact that producing additional energy on the surface of the Earth will raise its temperature. We will stop increasing our energy use one way or another once we hit this wall.
If we want to continue using more energy we'll need a whole second Earth to do it on. Great, we've colonized mars! What does that get us? Based on a 2.3% growth rate and the Rule of 70[3], we'll use up that second Earth in thirty years. We'll now need two Earths to keep growing for the next thirty years.
[1] https://dothemath.ucsd.edu/2011/07/galactic-scale-energy/#:~...
[2] https://www.youtube.com/watch?v=F-QA2rkpBSY&pp=ygUodGhlIG1vc...
[3] https://en.wikipedia.org/wiki/Rule_of_72
Globally rather than an exponential curve instead the global quality of life keeps rising as more people enjoy the benefits of modern technology like AC and tablets, but the number of people isn’t continually increasing. Birth rates keep declining so in the short term its populations catching up to increased lifespans.
But then you run into other problems. Can an economy like ours—which is wholly predicated upon unbounded exponential growth—continue indefinitely when energy use is effectively capped? Yes, there are efficiency gains and productivity gains to be made. Yes, our population growth is slowing and fill eventually flatline or even decline. Those will extend the length of time before we fully exhaust Earth's energy budget. Growth will end, and likely on a significantly shorter timescale than recorded history.
https://dothemath.ucsd.edu/2011/07/can-economic-growth-last/
https://dothemath.ucsd.edu/2012/04/economist-meets-physicist...
This is a false assumption. The economy is based on doing what people want. If people suddenly want glow in the dark kitchens someone will ramp up production of glow in the dark paint and take customers from companies that didn’t follow the trend. That’s the feedback mechanism keeping the economy functioning.
Growth at the micro level and growth at the macro level aren’t the same thing.
I absolutely would not confine ourselves to land as oceans provide a large area available for solar energy capture and there's no reason to think we might not be able to use it.
Photovoltaics, or whatever technology we use to capture sunlight, will get better but even at a modest 20% still gives us a lot of headroom.
Whether its 2.3% or 2.5% energy growth per year, the calculation gives us a timeline of 200-400 years. For some reason this is used as a countdown to oblivion instead of a rallying cry about where we're headed. Mars is great but there's a *lot of space in space*. Besides pushing solar panels into orbit, either the earth, moon or sun directly, there's also tons of asteroids, ripe for mining.
Forget a second earth, we can make a Dyson swarm. What you take as a countdown timer to a bomb, I take as a timeline for us to go up the Kardashev scale.
These reduction to absurdity arguments about the temperature of the earth assume we're not going to space. I don't understand why you reject this idea outright.
All the above calculations about the energy available to us are from a tiny pin-prick sliver of how much energy the sun deposits in all directions, every day, all day, for the 4 billion years. Once we have access to a significant fraction of the suns energy, going to other stars and repeating is well within feasibility.
Posts like abetusk's are a great illustration that "the solar budget" is a very generous energy budget. That may seem too obvious to mention, but in 20th century ecology literature (or even as recently as the early 2010s) living within "the solar budget" was often conflated with a low-energy, deindustrialized future. Constant growth fueled by sunlight (or anything else) can't go on indefinitely, but there's also no prospect that a sunlight-fueled world would have less energy available than the old fossil-fueled one.
> If we want to continue using more energy we'll need a whole second Earth to do it on
Or we make like Niven's Puppeteers and move the Earth out into a further orbit with less insolation.
Nuclear power, on the other hand, releases energy stored in fissile heavy elements that were produced by the death of previous stars. There is no natural process using incident light from the Sun to create them.