The question the paper asks is: if all of global electricity is generated using solar panels, will that have any impact on weather or climate elsewhere.
Their model: Put all of those solar panels in Northern Africa!!!
In this extreme scenario, yes, their are possible impacts. Since, such an extreme scenario is unlikely to happen, this is only an intellectual exercise.
> Their model: Put all of those solar panels in Northern Africa!!!
This is how you develop a model. Ask if there are effects in extremis. If not, easy answer. If there are, you figure out how to measure them and then deploy that metric on the hard stuff.
No. This is how you generate attention. The proposed dimensions are equivalent to several centuries(!) of the global solar panel production, just to get to such coverage.
It's just as ridiculous and removed from reality as the claim that tidal energy isn't sustainable and would tidally lock Earth to the Moon in 1000 years that someone published [1].
Spherical cows for the publish-or-perish paper mill that's modern academia is what it is.
The example is extreme, but they also point that out in the article. It's still useful to be able to model this, and as they point out, if you use a less extreme but still extremely ambitious 5% of the Sahara, the effect becomes negligible.
So in theory, solar power could affect the climate by itself, but it practice it won't.
But a model that shows that if you change something in this part of the world, it's going to affect rainfall in these other parts of the world, is still very interesting and potentially useful. It could help reduce desertification.
Trying to imagine the transmission capacity required to pull off routing all of the world's electricity from a single place in North Africa.
However, it seems like their model just shows a modest-ish heat island effect? It's not clear that the change in the albedo from even covering 50% of the Sahara in PV panels is much worse than the likely effects of climate change under the typical +2 degree C climate model?
They're still barely ankle deep in analysis to be honest.
Saharan sand has some not obvious from the outset effects on global climatic patterns (it's worth a look into).
More than 60 percent of the 1.5 billion tons of mineral-rich dust present in the earth's atmosphere comes from the Sahara. Individual dust clouds can reach a size of 500,000 square kilometers, an area equivalent to that of Spain. As these dust particles are only airborne for a short while, it's difficult to analyze the full extent of their impact.
It wouldn't make any sense to generate the world's energy centrally and transmit - economically, politically, etc.
The Americas could generate 100% of energy needed from <0.25% of land in Texas, Arizona, New Mexico, Nevada, California, Chihuahua, and Atacama.
And even that wouldn't make sense - because there's so many other places to generate solar, transmission is expensive, and you want your generation decentralized due to weather and many other reasons.
Why would China buy solar from Sahara when they have the solar radiation to get it domestically from Southwest Tibet?
There are numerous points of failure in any grid. the big power generators and transformers are hard to hide targets.
A big surface area of solar panel is much less dense in terms of power generation per square feet. So you would need a lot of ordinance to meaningfully decrease its output (except at transformer points, but that’s like in any grid).
They're not that hard to extinguish, you can just wait for them to stop burning on their own. Which is generally how car fires are also extinguished. The good news is EV fires take a long time to burn through the fuel, allowing anyone in the vehicle plenty of time to safely get out of the car.
Sure, it's a bit annoying for the fire fighters who turn up 20 minutes later... but I think even they would rather arrive to find a burning car if it means the people who were in that car are safely watching from a distance.
I still think that's an important question to answer, though. Yes, obviously we aren't going to pave an entire desert with PV, but it's worth asking what the consequences would be because we might someday, or might do something similar.
...an intellectual exercise which is an invitation for sensationalist reporting to misquote and for climate denialists/fossil fuel lobbyists to then use as a reason for opposing solar power.
Global energy use is ~20 TW. Say panels generate ~50 W/m2 average (about the value of my home solar in CA after all losses and time averages.) So, you need 20e12/50 m2 = 0.4e12 m2 ~ 400,000 km2. Area of Sahara = 9.2 M km2, so 4.3% of the Sahara to satisfy world-wide energy needs with just solar.
To give you an idea of the climate impact, according to the World Bank website, all urban areas on Earth total to ~2 M km2. So we have already built at least 5x the area of this imaginary solar array.
"And in any case, these effects are based on hypothetical scenarios. Our Sahara scenario was based on covering 20% of the entire desert in PV solar farms, for instance, and though there have been ambitious proposals, anything on that scale is unlikely to happen in the near future. If the covered area is reduced to a more plausible (though still unlikely) 5% of the Sahara, the global effects become mostly negligible."
All that, then this. Hard to take this seriously outside the realm of academic curiosity.
In academics, even a negative is useful information. This article could be reframed as "no need to every about changing the climate "
As they point out, even covering 5% of the Sahara, or huge swathes of the Aussie Outback, (both unfathomable large areas) would have negligible effects.
20% of the Sahara would be 331 terawatts of solar panels.
To put it in perspective that's an order of magnitude more than the entire global power consumption right now.
I mean sure, I guess one day we'll probably consume that much power. But it won't all come from one place. It will be diversified from solar panels spread all over the world as well as and countless other forms of energy generation.
Covering 20% of the Sahara in solar is a ridiculous scenario. And the impact would be, what, a slight weather variation in a few other places around the world? OK. Sounds pretty good actually.
20% of the Sahara is a lot of land. I think solar is a great addition combined with other things. My primary residence is in a rural county with a lot of cattle/range land where solar farms could help greatly with something like a microgrid. At a larger level we need wind, hydro and nuclear as well.
Not reasonable to write an article that sounds like this is an actual problem, only to tell people in the third last paragraph that it is largely a hypothetical scenario and in all likelyhood not going to matter for any real-world scenario.
Presumably solar panel arrays suspended above grass improve the local micro-climate for grazing animals below, and when suspended over canals they may also reduce evaporation.
The negative harms should be studied, but also balanced rationally.
We need to have a wider / deeper discussion about the relative harms vs benefits of solar "geo-engineeering" - putting particulates or salt water droplets into the air to reflect sunlight, to reduce warming.
This is urgent - the CO2 is already there, and we will likely have a long plateau of current highest emissions for another decade .. its the area under that curve that matters - total CO2 - and this will likely bake in +2C in coming years.
We have some data from vulcanic eruptions and recent reduction of particulates in shipping fuel, so we know in principle it works.
We've been releasing sulphites inadvertently as pollution for a long time .. it has not been world ending.
We need to do experiments and get better estimates on the benefit and harms, and feed that back into our climate models, and climate and weather predictions if we do larger geoengineering.
If we stopped all carbon burning tomorrow, we still need a way to bring the heat down - the ice and sea have buffered a great deal of the damage, but the ice will keep melting for a long time from current GHGs, perhaps irreversibly on human time scales. Id argue that harm is orders of magnitude greater than the negative effects of releasing sun-reflecting/shielding particulates.
To act or not, which does more harm ? That is a question science can give the best answers to. We need to get over the 'controversy', do the science we can in the short time we have and then act.
"If we build enough of X to a ridiculous level never before seen by mankind, will we have unexpected other effects? Our computer models (based only on levels seen before) say yes!"
Same, it had a considerable effect, which in turn caused some animals to be more prolific than others. I wonder what will the result be in a couple of decades.
If anything this might help future decisions to be taken more seriously.
Already are to some extent, my brother worked as an environmental geodata analyst for a big electrical landline project and they had biological surveys done for every meter of the whole planed route so the impact on potential wildlife and vegetation could be weighed. But the long term stuff is still under-studied
It looks like the simulated a truly massive installation, which they note is interesting but a bit unrealistic.
This seems, though, if anything (and, of course maybe it is nothing), to be an argument for not putting all of the panels in one desert. Which is fine. Actually it would be nice to put renewables everywhere so regions can be energy-independent.
There's more than one desert on the planet. Australia has a decently sized outback. The Southwestern US could accommodate a not-small chunk. Antarctica doesn't really need one, so that's 4 of the 7 continents. Europe and Asia are pretty far north, so might be harder there, but Chile's Atacama would potentially work as well for South America
Or, you could just put them on the roofs of existing buildings. I mean, that's where you need the power right?
All you need then is a grid so you can off load excess power (you're going to have a lot of excess power, if you cover even one in ten roofs in a city) to some kind of storage method and draw back from that when the sun isn't shining.
I would think the environmental difference between a ceramic tile or metal roof and a solar panel roof is pretty minimal.
Asia is pretty big. Arabian, Gobi and Thar deserts are not that much north. They are also near densely populated areas of the Middle-east, East China and Gangetic Indian plains.
> it would be nice to put renewables everywhere so regions can be energy-independent
Energy sovereignty is an underplayed reason for getting off oil & gas as quickly as possible. Why make your economy dependent on the whims of authoritarian states? Bad for national security.
It's a big reason why China is pursuing renewables. They are dependent on oil imports which can be easily blockaded if they ever decide to invade Taiwan.
I’m also hoping it can restore the economy in some rural regions. For example, imagine if they put a big installation in somewhere like West Virginia. Inexpensive property costs, close to big markets like the Northeast Megalopolis, maintaining solar panels seems like a way more pleasant job than working in a coal mine…
A small personal observation, from where I live (southern French Alps): when it's cloudy temperature is much less cool in the night, much less hot in the day, meaning clouds act as a blanket able to keep heat in the lower atmosphere, under the clouds, while similarly absorb heat from the Sun impeding it to reach the lower atmosphere.
Meaning: solar modules do absorb heat, but they are normally not ON GROUND, they are a bit lifted, so the heat they absorb during the day get quickly dispersed by the atmosphere during the night, they thermal mass is little, making days a bit more hot in the air, but similarly shielding the sand underneath to get hot, since it's shaded by the modules and that makes night a bit cooler since the sand have much more thermal mass. This circadian delta might accelerate air circulation, witch might move differently the whole earth atmosphere but simulating the substantial effects seems a bit too much for our modeling capacity, too many variables, too much possible drifts.
Floating solar seems to make good sense; the 'land' is otherwise unused and unusable, the water can be used for cooling, it offers shade and a structure for marine life. Of course cabling and storm-proofing etc means much more expense, but if we're talking about gigantic installations ...
Solar is very much about surface area. But you know how much weather happens in those two meters above the surface of the ocean? So we put wind turbines out there but not solar panels.
Hang weights from reels, underneath, for energy storage: reel them out to generate, reel them back up to store excess for later. Gang dozens of reels on each motor/generator, for economy. (Using just one reel per motor/generator is what makes mineshaft energy storage uneconomical.) Make the weights out of peridotite rock, which is very cheap and 60% denser than most.
Certain places are better for this than others. Offshore of Rhodes, eastern South Korea, Monaco, the Philippines have very deep water not far off shore.
Or... drop an anchor, pull the floating barge against the tide towards the anchor when the sun/wind are optimal. Let the tide pull the barge away from the anchor (with a generator on the winch cable) when you want power while the sun/wind aren't providing power.
Tides have massive potential for power generation. The best part is how predictable they are.
Your idea of weights being raised and lowered makes more sense for land based energy storage (and it is being trialed right now, with promising results).
It is very hard to share a motor/generator between multiple weights on land, because you would need multiple adjacent mineshafts, so the economics are difficult. At sea there is plenty of depth anywhere you care to sink something.
> For instance, huge solar farms covering much of the Australian outback would make it sunnier in South Africa, but cloudier in the UK, particularly during summer.
Given how much the Brits whinge about anything over 24°C, as an Australian I think it's our duty to the world to do this.
a useful piece of context here is what it would mean to cover 20% of the sahara with solar farms and why the authors think it's unlikely
i haven't read the paper, so i don't know what fill factor they're considering. say it's 50%, so the spaces between the panels occupy as much area as the panels
9.2 million sq km · 20% occupancy · 50% fill factor · 21% efficiency · 29% capacity factor (which is probably a bit low) · 1000 watts/m² works out to average production of 56 terawatts, roughly three times current global marketed energy consumption, which is 18 terawatts
that is, building such a thing involves quadrupling the energy budget per person. right now the humans use about 18 terawatts, about 2.25 kilowatts per person, roughly equivalent to the labor of 22 workers. we're talking about boosting that to 10 kilowatts, so the average person has the equivalent of 100 laborers at their disposal, so it will be a vast reduction in the cost of energy
and the sahara isn't the only place you can put solar panels, even on earth, even on land
what would a human society that's four times as energy-intensive look like
they also say that if you only cover 5% of the sahara in solar panels, so that this one power plant produces more energy than the entire human race does today, 'the global effects become mostly negligible'
— ⁂ —
in the other direction, consider their opening declaration, 'the sun's energy is effectively limitless.' the sun only outputs 380 yottawatts, which is only greater than the 18-terawatt number above by a factor of 21 trillion. so human energy consumption can only grow by 21 trillion times before finding other, larger energy sources
okay but is that a lot or what
if we take the historical post-industrial-revolution growth rate of 5 percent per year and apply it to energy consumption (perhaps on the assumption that there's a lower limit to how much energy we can dissipate per unit of value produced, even if we haven't reached that limit yet) that's only 629 years of growth. that's longer than a human lifetime but still an eyeblink in the two-million-year story of humanity or the ten-thousand-year span of recorded history
if, by contrast, we start building von neumann replicators that mine mars and venus in order to double their population and energy consumption on average every year, it's only 44 years, and many people still alive will see it happen
so i think 'effectively limitless' is shortsighted
Yeah, solar panels reduce albedo. But isn't the idea that this reduction in Albedo is far more than counterbalanced by the reduction of greenhouse effect it can bring?
This could also work as a benefit --- for example, the 'heat island' effect of (sub)urban areas. True, the panels are black, but now A/C (and other sources of waste heat) is running on local energy, not energy fed in from far away.
I have no idea how significant the effect would be, but it would be an interesting study to do / read.
So there could be a causal relationship between solar installation and local weather of the installation. TBH this had never occurred to me, so its certainty and interesting thought experiment.
Mind wanders off to solar installations in space which pops up in mainstream press a few times a year.
I do think about wind turbines slowing down earth’s rotation. I’ve not done any calculations but I’d guess even a gargantuan number of them would have no noticeable effect.
The only way to change the earth's rotation is to transfer some of its angular momentum to something that's not part of the earth. The atmosphere is part of the earth, so just interacting with it can't do anything.
With the same angular momentum the Earth's rotation still varies. Snowfall high in the mountains slows the rotation by a minuscule amount. Mountain glaciers melting speed it up slightly. But again, negligible amounts.
Curious to be proven wrong, but I’d imagine they’d have no net effect.
The only thing I could imagine is a miniscule gyroscopic effect if the turbines were aligned, but that probably nets out through tectonic weirdness before altering Earth’s rotation. Maybe if the slower-moving air sinks on average and lowers the planet’s moment of inertia? (Though that would speed up the Earth’s rotation.)
Yeah, the author went through some extraordinary mental gymnastics (i.e. keeping exponential energy consumption growth steady over centuries) to get his calculations to line up. It's almost grotesque how our species manages to vastly underestimate its influence on our planet in some aspects while simultaneously overestimating it in others.
White ones would be less efficient because the white color consists of reflected photons, and there's a lot of energy in those photons that a modern panel would be able to convert into electricity.
The revenge of captain obvious. There is no free lunch. (Yes I know this will come at the cost of some downvotes). In general, we are not only bad but hilariously bad as a species in thinking in 2nd and 3rd order effects.
Next up: gigantic wind farms change weather.
Further up: we need to reduce our energy consumption, or vent excess heat directly into space. Give it another 10 years or so. CO² is a relevant and dominant factor in global warming, but I'd bet a lot of money that it isn't the only one, just the dominant one we know of right now.
By the way: the amount of surplus heating currently observed in the oceans exceeds the predicted amount by around two orders of magnitude. Too tired to provide (indirect) sources. At least this was my understanding by following the traces of for example this article¹.
Of course, digging out oil, and dumping all of the shit into buried ditches, and then setting the oil on fire has no affect on the surrounding environment.
Their model: Put all of those solar panels in Northern Africa!!!
In this extreme scenario, yes, their are possible impacts. Since, such an extreme scenario is unlikely to happen, this is only an intellectual exercise.
Paper link: https://www.nature.com/articles/s43247-023-01117-5