Modeled impact of large-scale wind and solar farms in the Sahara | Hacker News Reader

Modeled impact of large-scale wind and solar farms in the Sahara

phys.org

102 pointsjoeschmoe3111 comments

> The wind and solar farms simulated in the study would cover more than 9 million square kilometers and generate, on average, about 3 terawatts and 79 terawatts of electrical power, respectively.

This study is looking at a combined farm size that's almost as large as all the land mass in the US.

https://en.wikipedia.org/wiki/List_of_U.S._states_and_territ...

That seems a bit excessive, considering that the world energy consumption is only about 18 terawatts (and only about 3 terawatts of that are electrical right now).

Civilisation is to a large extent constrained by available energy, which is why new energy sources have lead to industrial revolutions. As long as you're not spending other resources on it excessively any massive amount of energy surplus can be put to good, previously uneconomical and fantastical-seeming uses.

Given enough energy you could regulate the entire biosphere by brute force.

And if you ever want to get serious about colonizing space a stupendous amount of energy has to be spent just lifting people and materials into orbit.

I totally agree with you, but if any of the many variations of space elevator becomes possible, you can balance the energy and momentum of mass moving up, by mass going down perhaps returning tourism or valuable materials mined off earth

Which is also awesome, but not an argument against installing massive, terawatt capacity photovoltaic solar arrays across a desert.

did it occur to you ... perhaps I wasn't arguing against terawatt capacity PV solar arrays across desert? ;)

If countries like China, India, Malaysia, Nigeria etc. start to have per-capita energy use close to that of the United States, global energy consumption could easily double or triple even without any increase in population.

I said "if", but I really should say "when", because this is what's expected to happen as these countries modernize and start getting things like air conditioning on the same scale as US households. 79 TW of capacity will become a necessity sooner than you think.

Almost no one in Europe has per-capita energy use anywhere close to the US, so why would China? And I don't think it's just that we have less AC, though that's surely the strongest contributor.

For reference for anyone curious: the US has a final energy consumption of 6.0kW per capita, the EU average is about 2.9kW. (2009 data)

At that point, China was at 1.3kW/citizen, Africa at half that, India on the level of Africa and the middle East slightly below the EU with 2.3kW.

Wikipedia [2] uses world bank data instead which seems to measure differntly, but the relations are roughtly the same.

1: https://www.eea.europa.eu/data-and-maps/figures/final-energy...

2: https://en.wikipedia.org/wiki/List_of_countries_by_energy_co...

I'm confused by your wording and by wikipedia. Watts are a measure of power, not energy. kWh is a measure of energy.

In the same sense as your server has an "energy consumption" of 2000W, i.e. 1W <=> 356*24/1000 kWh/a.

By saying they use 6kW I mean that each year they use 6kWa (kilowattyears), or 52569 kWh. And 6 kilowattyears per year is the same as 6 kilowatt.

The main issue is that Americans have much more house than Europeans do. The three-car garage McMansion with double-height ceilings and a double-height foyer is a ridiculous amount of space to cool down and heat up for a single household.

Nice visualization of that kind of data: https://www.reddit.com/r/dataisbeautiful/comments/7ei5f4/com...

>The three-car garage McMansion with double-height ceilings and a double-height foyer is a ridiculous amount of space to cool down and heat up for a single household.

That's definitely not a significant contributor to American energy usage, given that not everyone in America is a millionaire.

There are 139 homes for sale in the Dallas Metroplex that have 6 or more bedrooms for $600,000 or less.

There are 1.318 million Dallas residents. 139/1.31M = 0.0001. Not too significant.

That's not that amazing of a metric. For starters, households are what track to a house, not residents. Then you also have to consider what proportion of that population is looking to buy a house at any given time; most people tend to stay in a purchased home for a long time, so they may already live in a McMansion style house.

McMansions are McMansions because they are cheap to construct and cheap to buy (and surrounded by thousands of nearly identical structures).

We have less AC, but we have more heating, which is more energy intensive than AC.

I’m doubtful about both of these claims (Europe has more heating than the US and that heating is more energy-intensive than AC). The latter seems particularly hard to believe given how efficient heating is compared to A.C.

Sure thing. Heating (space + water) represents nearly 80% of EU household energy use[1], compared to 60% in the US[2]. I have no clue what you mean by "heating is efficient compared to AC". In fact, opposite is true -- unless you use some kind of heat pump, which are very uncommon in Europe, you only get exactly one unit of heating per unit of energy consumed. On the other hand, with AC (which _is_ a heat pump), you get something on the order of 5 units of heat (energy) removed per unit of energy consumed.

[1] - https://ec.europa.eu/eurostat/statistics-explained/index.php... [2] - https://www.eia.gov/energyexplained/index.php?page=us_energy...

This way of thinking about efficiency doesn't take into account gains from insulation and/or reducing unnecessary heating. An electric blanket is more efficient than heating a room, which in turn is more efficient than heating a house.

Here is a study from 2013 in Environmental Research Letters. Comparing Minneapolis and Miami, the coldest and the warmest metropolitan areas in the US. Minneapolis uses (per capita) 3.5 times more energy for heating and AC (mainly heating) than Miami (mainly AC).

"This finding suggests that, in the US, living in cold climates is more energy demanding than living in hot climates."

http://iopscience.iop.org/article/10.1088/1748-9326/8/1/0140...

Part of the difference is that, heating is usually directly burning oil or gas, so the efficiency is 1:1, whereas cooling can have efficiency 4:1 because heat pumps just move heat. But you need electricity to run a heat pump, and generating the electricity is at about 0.5:1 efficiency. So if all Minneapolis switched to heating with heat pumps, they could bring the difference down from 3.5 to 1.8. But still living in Miami is more energy efficient.

Miami can only be considered "hottest" because it never gets cold and you consider the average. It actually spends a long time near room temperature. Somewhere like Las Vegas or Phoenix gets both hotter and colder.

If we arbitrarily pick 70F as room temperature, Miami has a typical annual high/low spread of approximately 60F-90F or -10 to +20 degrees relative to room temperature. Meanwhile, Minneapolis has a typical annual spread of approximately 10F-85F or -80 to +15 degrees relative to room temperature.

You should look at section 2.2 in the article. Miami has more cooling degree days (2423) per year than Phoenix (2327) or Las Vegas (1786).

My main intention when I started responding was to point out the much larger winter heating differential in the north, in response to the discussion about heating versus cooling costs and efficiency. I am not sure why I got distracted about Miami versus Phoenix and inserted that before posting.

My experience living for a few years in a tropical country was that locals acclimate and do not cool nor dehumidify their living spaces nearly as much as many Americans seem to do. Much like in Florida, you will see people running around in jackets or even knit hats on slightly cooler days when someone from a colder climate would already think it is warm and time for shorts.

I think it is difficult to compare different regions with different climates and cultures. It seems impossible to me to choose a metric that isn't inherently biasing the analysis towards one arbitrary normative standard. Compare similar regions or one region year-to-year to evaluate the efficiency of local practices.

Edit: another issue is the daily cycle. You can use thermal mass to smooth out daily temperature extremes but that doesn't work when you spend weeks or months with temperature differentials that remain offset from comfortable.

Heat transfer, though, is based on temperature differentials - if Miami is +7degrees every day of the year, and Phoenix is +20degrees only 117 days per year, the energy demands of Phoenix for cooling could very easily be much higher (energy lost through walls, roof would be much higher, as would mass transfer of energy when doors open.) This is irrespective of local construction. Add to this, radiative gain (insolation through windows) is not captured in the degree-days statistic, nor is humidity related cooling (condensing water takes a lot of energy!)

No. No no.

Heat conduction rate is linear in response to the temperature difference [1], so 1 day of 20 degrees difference should give the same total heat loss as 2 days of 10 degrees difference. They also both give 20 degree days, so degree days is the correct measure and should correlate linearly with the energy needed for heating or cooling. The same goes for the heat content in the air that is exchanged due to draft, opening doors etc.

You might have a little bit of a point with sunlight, and humidity. But my guess is that they don't dominate, compared to the conductive and convective heat exchanges.

[1] https://en.wikipedia.org/wiki/Thermal_conduction#Fourier's_l...

Right, thank you for the correction. I was mistaken about the conduction. I'm not sure why I was thinking it was something other than linear. So degree days should account fairly closely for the conductive heat transfer through walls.

I think my comment about mass transfer when opening doors could maaaybe be defensible but I'll leave it alone :)

I don't have the numbers. But intuitively, AC should be less energy as it's energy transfer. Heating is usually directly adding energy to the living environment. You can get heating using heat pumps (some southern US homes have this type) but it doesn't work very well when it's too cold outside. Europe should have more heating than US, with it's more northern location.

Indeed. Heat pumps routinely have efficiencies[0] around 400%-500%, because they're not really "producing cold", they just move energy from one inside to outside (or the other way around).

[0] - of course that's a specific meaning of "efficiency"; they're still less than 100% efficient in terms of energy expended on moving other energy around vs. theoretical minimum.

Electric heaters are more efficient than traditional air conditioning units. Transferring heat from inside a 80F house to the 110F air outside is not very efficient, only about 30-60%. On the hand, an incandescent light bulb is a ~99% efficient heater. The only energy that is "lost" is the light which is radiated out the window. Space heaters and electric radiators are 100% efficient.

Additionally, heat pumps are can be spectacularly efficient when it's warm outside (in which case, who needs a heater) but are still more than 100% efficient when it's cold outside. Waste heat is still heat, after all. The problem with heat pumps is that the quantity of thermal they can put into a house drops dramatically when it's cold outside. When it's extremely cold outside, the only heat you're added to the place you're trying to heat is the waste heat from the unit itself.

Swamp coolers can be very efficient "air conditioners" but they only work in extremely dry environments. Heat exchangers that pull cold water from a nearby cold lake (for instance, there are data centers in Chicago that use the lakewater from Lake Michigan) can be highly efficient, but there's a lot of infrastructure involved, and they only work on warm land near cold water.

The coefficient of performance (efficiency) of an AC at that temperature range would still be well over 1. Probably around 3-4 depending on the unit and refrigerant. [0] ACs are very efficient for the same reason heat pumps are efficient. They are just moving energy around.

[0] https://www.researchgate.net/publication/242172797_A_Compari...

> Transferring heat from inside a 80F house to the 110F air outside is not very efficient, only about 30-60%.

Where did you get this number from?

> Transferring heat from inside a 80F house to the 110F air outside is not very efficient, only about 30-60%.

This is wrong. Modern heat pumps can easily have efficiency over 400%.

It may have to do with where the heat comes from? I live in southern most areas of US, but from what I understand it's common for heating to use Gas/etc. instead of electricity.

Therefore they use less "energy" (i.e. Electrical energy)

All just a theory

The most efficient way to heat would still be generating electricity from gas. You lose 50% of the energy (well, you can still use part of the residual heat for heating, too), but with the electricity, you can run a heat pump that heats with about 400% efficiency. 50% × 400% is still 200%, so twice as good as burning gas for heat.

Exactly,

As we see efficiency increase as well.. things are not as bad. Sure energy consumption will increase, but not to US levels.

We should expect increases in efficiency vs the US to /increase/ energy use relative to US levels.

https://en.wikipedia.org/wiki/Jevons_paradox

Demand for AC will increase greatly, no?

Also, if that much excess would otherwise be wasted, we could always use it to smelt aluminum, form it into big tanks, attach standard docking rings, fill them up with water, and launch them into a designated band of refueling orbits. Those would almost certainly be useful later in space technology development.

Or condense air and harvest co2 for sequestering.

I’d like to offer a comparison that may put it into a different perspective for you. Imagine saying that a 1 TB hard drive is excessive, 20 years ago. Sure, contemporaneously, but we sure did find some useful ways to put all that excessive hard drive space to use.

One really good example would be water desalination. I'm sure the neighbors of the Sahara could use a glut of fresh water.

A great use for the power, but also another destabilising thing for them and all the interested foreign powers to fight over.

Unfortunately I think you're probably right. It's really a shame how many otherwise great things we miss out on because people suck. Constant "this is why we can't have nice things."

Well, the extra energy can be used for CO2 capture :)

And the win of an arable Sahara would be just incredible.

Generating terawatts is not unreasonable. But transportation of it out of North Africa would be almost as big an effort as building the solar farms in question.

How would you go about it? Lots of high-voltage lines, or is there something else that'd make more sense for this unusual setup?

Distributed generation seems like it would make sense (with current tech). Italy and Spain could help here.

Maybe you could go from Tunisia to Europe, but supplying African countries seems more attainable.

It'd be nice if you could turn the energy into a gas like methane, pull CO2 out of the atmosphere in the process. Good for the atmosphere and you could transport gas over long distances.

https://en.wikipedia.org/wiki/Power-to-gas

I wonder if hydrogen is practical to transport in large quantities. My guess is no.

Your idea of methane, if it could be done efficiently, seems more likely.

Ammonia might be a better way. Easy to produce, store and transport and contains higher hydrogen density than compressed hydrogen (I think).

http://www.ammoniaenergy.org/

The biosphere can't handle ammonia accidents at anything resembling the scale of crude oil and natural gas accidents. Methane is the better choice.

I saw that very recently an Australian company had worked out a way of transporting hydrogen in large quantities by converting it into something else which I conveniently can't remember.

It was the CSIRO converting hydrogen to ammonia for transport, and then back to hydrogen again: http://www.abc.net.au/news/2018-08-08/hydrogen-fuel-breakthr...

Standard 800 kV DC line from Algeria to Berlin (2400 km) would incur less than 10% transmission loss.

What's the maximum current carrying capability for this standard 800 kV DC line?

Not sure for a standard line but the record is "1100 kV voltage, a 3,000 km (1,900 mi) length and 12 GW of power". So presumably 11,000 amps or so for that one. I guess you can have as much current as you want by making the cable thicker.

Being built in China here approx https://www.google.co.uk/maps/dir/Guquan,+Putuo,+Zhoushan,+C...

It's quite an impressive looking thing (vid) https://www.youtube.com/watch?v=VZ2KfrP_R3s

so transporting 1 TW would be a hundred of those in parallel

Yeah, but the average EU power consumption today is 0.35 TW, so with only a hundred of those in parallel (more likely spread around the Europe) you could literally satisfy 100% of EU power demand, and still have plenty of spare capacity.

I guess something like that. Probably beefier pylons and more cable on them.

could submarine power cables in the Medeterranian sea be thinner since the water keeps the cable cool?

Dunno - bit beyond me that one.

I suppose that a lot of such transmission lines should be superconductive, if only to help put enough electricity through physically more narrow channels.

Even with modern high-temp superconductors, it's a lot of expensive infrastructure, a lot of protective redundancy, and a lot of maintenance.

> Even with modern high-temp superconductors, it's a lot of expensive infrastructure, a lot of protective redundancy, and a lot of maintenance.

Which means lots of local jobs, which may not turn out so bad in the short term.

Why should it be transported out of North Africa? That huge power availability would be a much needed incentive to attract factories there. With so much energy, water can be desalinized and transported near the generation sites so in the long term the energy would transform the desert in a very desirable location.

At least 600kv HVDC lines for moving gigawatts. Several billion US dollars to build. Might be worth it to tie into Europe's grid.

That's a lot of terawatts. In 2013, the world used about 18 TW on average.

https://en.wikipedia.org/wiki/World_energy_consumption

In this case, "large" means almost the entire Sahara, which is 9.2 million square kilometers.

Africa is huge. 79 terawatts of power in the study is not unreasonable, given the population growth.

https://en.wikipedia.org/wiki/Sahara

I'm quite sure the availability of cheap energy would result in quite a boost for most countries around this complex. As a result their energy consumption would most likely increase a lot.

So in the Sahara that would probably mean all solar and a few batteries. Wind turbines would have maintenance issues from the sand storms.

I would not have guessed that the solar arrays beneficial effect on precipitation is dependent on them not being too efficient and so releasing more heat than bare desert (which is very reflective).

From the article - "The precipitation increase in our solar farm experiments is due to the relatively low conversion efficiency of the panels (15%, typical current conversion efficiency for photovoltaic panels), which results in albedo decrease. However, if solar panel efficiency and the associated effective albedo are high enough to lead to an albedo increase relative to the background environment (as, for example, a 45% efficiency would), the climate impact would be surface cooling with precipitation suppression, similar to the impact of overgrazing in the desert. Assuming an intermediate conversion efficiency higher than 15% for solar panels (e.g., 30% efficiency) results in negligible albedo change and, thus, insignificant climate impacts."

[^] http://science.sciencemag.org/content/361/6406/1019

I still consider it a win even if solar panels reach 45% efficiency: surplus energy can be used to desalinate and transport seawater for irrigation, or by condensing it from air (releasing the heat of condensation)

>"The greater nighttime warming takes place because wind turbines can enhance the vertical mixing and bring down warmer air from above," the authors wrote.

That's actually a very interesting effect. I guess the votices generated by wind farms quite beneficial in a desert region.

It would be interesting to see what effect wind farms have on desertification. Given that two large factors are fertile ground blowing away and desert sand being blown onto fertile land, I imagine wind farms might be beneficial for slowing down desert growth.

I guess the votices generated by wind farms quite beneficial in a desert region.

Unless you're an ectothermic creature (like a lizard, tortoise, snake, etc...) which relies on the cool of night to hunt food and find water.

Yeah, I was thinking that the cool temps at night are necessary for condensation. Without it, the plant and animal life in the region could be devastated.

If we're talking about deserts, there is next to no plant and animal life, and part of the goal of this would be to change that.

You are grossly misinformed about deserts.

Humans tend to know less about environments that kill us rapidly.

While interesting, I have my doubts as to the reliability of any prediction model for a case such as this. It would be a case quite a bit outside the historical data which the prediction model is based on, and it's not like weather and ecological systems are nicely linear and well-behaved systems to predict.

What they do not model is the local effect of the cities that will need to be built to service all these panels. 9,000,000 square KM is a gigantic area. Millions of people will be needed to maintain these farms. Those people will need to either live locally or be transported regularly. The study should model how the presence of those people in such a relatively underpopulated area will impact local climate.

1 person per km2 = 9million people. But I'd bet you would need a couple dozen people for every km2 of solar farm, and associate infrastructure, meaning perhaps 100 millions people. And then all the people to provide services to those people ... this is like building a few new countries from scratch.

This line of thought interested me so I went and looked for some numbers.

The 10 sqkm 648 MW Kamuthi Power Plant [1] was built in record time in 2016 at a rate of 11 MW/day with a crew of 8500, which comes to about 2 months. This project is many orders of magnitude larger, lets be generous and say there is a 20x economy of scale in this project above Kamuthi: 100 sqkm / mo / 8500 crew. With 100 such crews (a lot) it would still take 3/4 of a century just to install 9M sqkm, and the installation alone would employ ~1M people. Wow.

According to [2], the degradation rate varies between 0.5%-1% per year. Using a generous 0.1% as an estimate for maintenance would put it at 9000 sqkm needing maintenance/replacement every year which according to the estimates above would only require ~10 full-time crews from above. Still ~100k people, a big city for sure, but not millions.

Another thing to consider is that a project of this size would be decades in the future and I imagine it would heavily employ automation, so dropping a couple zeros off the number of people required could be quite feasible.

[1]: https://qz.com/india/854483/worlds-largest-solar-plant-kamut...

[2]: https://medium.com/@solar.dao/everything-you-need-to-know-ab...

1% per year? So some bits of these solar plants will last a century or more? I'd put a 20-year expectancy on everything, to cover both degredation and periodic upgrades. That is 4%+ per year. Nothing functional lasts for a century in the desert.

That figure is for the power generated only. Exposing the panels to the elements is likely to cut the life of some of them short.

But all in all it's essentially a rock generating electricity in certain circumstances. Not much to go wrong here.

Solar utility installations in California that I’m familiar with employ fewer than 1 person per square km.

And how do they keep all those panels clean? Who maintains the transformers when they break? Who swaps out the panels as they reach their life expectancy? It isn't just the night watchman, but the entire community of people involved in any large outdoor installation.

Like any utility, you don't need a guy just hanging around waiting for the transformers to explode. You truck him in every 50 years when that actually happens. It's time domain multiplexing, essentially.

The cleaning is largely a solved problem too.. You don't need 50 $10k/year people with squeegees, you just need one person in a $50k truck.

https://www.alibaba.com/product-detail/China-Road-tunnel-wal...

If there’s one job that robots are certainly going to take, it is solar panel squeegee guy.

And how do they keep all those panels clean?

Robots, like these from Ecoppia:

https://www.ecoppia.com/wp-content/uploads/2016/08/Product_D...

https://www.youtube.com/watch?v=VIH-C29N82o

Do they have problems with security? Try putting down any large piece of infrastructure in Africa - and careful that it doesn't get carried away.

Can the energy used to water desalination from the sea ?

"A new climate-modeling study finds that a massive wind and solar installation in the Sahara Desert and neighboring Sahel would increase local temperature, precipitation and vegetation. Overall, the researchers report, the effects would likely benefit the region."

Beneficial to who or what? Beneficial in what ways?

You can continue reading the article to learn the answers, you know.

Sorry, I think the point I was trying to make was, do we really want to transform this unique geography? I live in Utah and if someone came here and said, "you know all this red rock and sand in the south, we could change that into a lush green landscape," there'd be massive opposition to that idea.

The Sahara desert is in fact growing (some 10%)... likely due to global warming. So it's constantly transforming in any case.

https://www.livescience.com/62168-sahara-desert-expanding.ht...

Do you think there would a massive opposition if someone came here and offered changing 2% of Utah into lush green landscape? I believe this change would be welcome, as people like some variety. Now, area of Utah is 2% of Sahara's area, so we could literally transform some patch of Sahara of the size of Utah, and still have 98% of it untouched, and get enough electricity to power most of Europe.

You haven't followed the recent Bears Ears controversy, have you? :-)

Here's another example. I grew up in Washington state. We have dams on a lot of rivers in the Northwest generating clean renewable energy before anyone had ever heard of climate change (Grand Coulee is the most famous). They literally did change the barren desert of Eastern Washington into a fertile farming breadbasket with the irrigation water and power they provide. Despite that, there are people who hate them because they changed the rivers into a series of lakes and are threatening the wild salmon population.

Pick any part of the Sahara and transform it and some species unique to that area, some native people, some curious weather pattern, some unique geologic formation will be lost and people will be upset.

I'm quite familiar with the dams, I live in Washington myself, and visited them multiple times. I agree that someone will always be upset, but this is generally true about every change ever. The point is to do changes that are most beneficial to the most, don't carry extremely serious side effects, and are net win over the alternative, which in this case is keeping burning coal. There's still plenty of coal left in Roslyn fields, and if not for the dams, it would probably continue to be extracted today.

Surely that coal is as big an argument for Saharan solar farms as it is for Washington dams?

Not really, though I imagine you'd be able to find 2% of Utah that nobody will care about in particular. I love Utah as much as anyone, especially in winter, but I wouldn't want all of the US look exactly like Utah -- and the whole US is about as large as Sahara is.

It is in fact still a matter of active debate as to whether or not the Sahara is natural. Some evidence suggests humans may have played a role in its expansion to current size, at least: https://www.sciencedaily.com/releases/2017/03/170314111320.h...

Question is, what do you care more about - the unique geography, or people living there?

Very few people live in the Sahara. However, to your larger point, many people live in Europe and the Middle East, and being able to power their cities could in turn obviate the need to emit quite a large amount of CO2, potentially affecting the entire planet in a positive way.

I would give up some portion of the Sahara (up to all of it) if it meant the planet doesn't warm 4 degrees. Of course that's easy for me to say -- I don't live there. :/

I have always been alarmed that I've never seen the weather effects of solar and wind power discussed - they are by definition removing energy (randomly) from the global climate system. As one of the least understood and most complex systems that _drastically_ impact our lives, this should be cause for concern.

It's not like the sun runs out of shine. I think the impacts of renewable energy is several factors lower than our current C02 based economy

that energy loss to solar and wind power is a miniscule percentage of the energy available, which is provided by the sun and most of which gets radiated away anyways. no need to worry about it for another 5 billion years or so.

The concern is not the output of the sun, but energy moving in an unplanned (and unplannable fashion).

In other words, solar panels shade ground that would have been sunned. Windmills change atmospheric differential pressure by removing windspeed.

The best case scenario is these changes are negligible, but that is a luck-based result. We have no idea because global climate is not effectively modelable.

You're worried that we haven't discussed the results of something that you admit is not effectively modelable?

Yes.

Just because something is difficult to model doesn't mean the outcome/results can't be wildly dangerous. We should still try and guess the impacts of our actions, but it seems like no one is concerned.

Reply on news.ycombinator.com