6 ms·

What storage and transport problem? This is a region that has a lot of sun, a lot of empty land near cities, and a lot of electricity usage for air conditioning

What storage and transport problem? This is a region that has a lot of sun, a lot of empty land near cities, and a lot of electricity usage for air conditioning. They can install a lot of solar before they need to worry about storage and transport.

kragen 6y ago
Maybe the problem of transporting the energy produced there to China or the Netherlands or something? The problem with selling energy in Morocco or Egypt is that the people there are, by and large, poor, so they can't pay you very much.
- jillesvangurp 6y ago
  Easier to generate it locally, which people in both China and the Netherlands do.
  - kragen 6y ago
    Indeed, and China has the Gobi, but the Netherlands' total solar resource is only a few terawatts, while hundreds to tens of thousands of terawatts are potentially available to be imported. https://www.en-tran-ce.org/custom/uploads/2019/02/Renewable-Energy-January-2019.pdf https://www.en-tran-ce.org/custom/uploads/2019/02/Renewable-... seems to say that last January the Netherlands had 4300 MW(p) of solar PV generating capacity, which produced 0.4 petajoules. If that's correct, the capacity factor of solar PV in the Netherlands averaged 3.5% that month, which is shockingly low, about three times lower than even Germany and nearly ten times worse than equatorial desert solar power plants — although January is one of the worst months, and eyeballing the graphs it looks like the year-round average is closer to 1.3 PJ per month, giving a deplorable 10% capacity factor similar to Germany's. Until and unless panel prices drop dramatically again, situating the panels somewhere sunnier will produce three times as much energy, enough to pay for constructing quite a bit of energy transmission infrastructure. But what happens if panel prices do drop dramatically? Well, if this appalling 10% is a typical capacity factor for utility-scale PV in the Netherlands, the total solar resource is only 4.2 terawatts, of which about 900 gigawatts is available with typical 21%-efficient solar panels. This is an order of magnitude higher than the country's current energy demand, but of course that means that you'd have to cover about 10% of the countryside with solar panels to fulfill that demand from PV alone — and seasonal variability means you'd have to triple that if demand is constant year-round. Covering a third of the Netherlands with PV panels is probably not possible, and covering more than 100% of it is definitely not possible. So the new kinds of energy-intensive industry that will become possible if PV panels get cheaper still will be out of reach of indigenous PV production.
    - jillesvangurp 6y ago
      These back of the envelope calculations are nice. But the underlying assumption of having just one source of energy to cover all needs is wrong. In the Netherlands we have lots of untapped potential in the form of wind, tidal, geothermal and imported hydro-electrical (from Norway). There are only 2 coal plants left (a lot of them were shut down in recent years) and quite a few gas plants and 2 relatively small nuclear plants that are very old at this point (i.e. likely to be shut down at some point). Between all of those sources, I suspect that importing solar from north Africa or China is not going to be attractive from a cost or efficiency point of view any time soon. Offshore wind is getting a lot of investment and dropping panel prices are going to cause a lot of homeowners to put these on their roofs. Lots of people put panels on their roofs already, which in most cases does not fully cover their needs but lowers their energy cost. I wonder if that is included in the report. Reports like this tend to focus on grid providers. A lot of solar generation is happening off grid. Home owners, businesses, EV charging stations, etc. are all use cases where solar panels are commonly used to lower energy cost.
      - kragen 6y ago
        As happened last time I conversed with you, you seem to have reinterpreted my words to mean something that is obviously wrong so that you can disagree with it, and then downvoted my original comment when I had the temerity to rebut you — this despite the fact that my comment explicitly disclaimed the absurd interpretation you imposed on it. — ⁂ — I didn't make an underlying assumption about having just one source of energy to cover all needs. I made an extensively-sourced, well-substantiated calculation that, out of all the sources of energy available in the Netherlands, two of them represent a large enough resource to support an order-of-magnitude increase in energy usage — enhanced geothermal and solar — and neither of these will be economically competitive with importing solar energy from elsewhere. The absurd misinterpretation that underlies your comment is that I'm talking about what will happen in, say, 2021, 2022, 2024, or 2028. But it should be obvious that the Netherlands is not on a path to use an order of magnitude more energy in 2028, nor will solar power in Morocco and the like usher in "new kinds of energy-intensive industry" that quickly. So, as should have been obvious from my initial comment, I'm talking about a longer timescale than that; I think 2040 to 2060. You chose to ignore this and instead interpret my comment as patent absurdity. — ⁂ — Let's talk a little bit about this "lots of untapped potential" and "a lot of investment" and "a lot of homeowners" who "put panels on their roofs" you tout in your comment. — ⁂ — First, what about the homeowners? Well, putting panels on your roof doesn't improve your capacity factor relative to the utility-scale plants surveyed in the reports I cited; in most cases it worsens it, because it's too expensive to reorient the roof to the optimal angle, and most people just use whatever the roof's existing angle is instead of building an extra truss to support the panel. So a square-meter 200-watt (peak) solar panel on a Dutch roof is still going to generate only about 20 watts on average, maybe 15 watts if your roof is at an unfortunate angle. Moreover, there isn't nearly enough roof for rooftop solar to meet a significant fraction of demand. Current marketed energy consumption in the Netherlands is about 60 gigawatts electric, 100 gigawatts total, about 5.6 kilowatts per capita — about 280 square meters of panels per person. Nevertheless, the total solar resource in the country is, as I said, 4.2 terawatts, of which 900 gigawatts would be available with current mainstream solar panels. It should be apparent that Dutch residential rooftop solar can only ever be an insignificantly tiny fraction of even the total current demand. It is an irrelevant distraction. — ⁂ — Of the other sources you mention, three — wind, tidal, and hydroelectric energy imported from Norway — are also insignificantly small compared to the country's solar resource. The fourth, geothermal, represents a larger energy resource than the solar resource, but because it is more expensive to access, it will remain untapped until after the Netherlands starts to experience a scarcity of places to erect solar panels. David MacKay's Europe-wide estimate of the wind power resource per person is 0.4 W/m² http://withouthotair.com/c30/page_232.shtml http://withouthotair.com/c30/page_232.shtml, which works out to 17 GW in the Netherlands' 42865 km². This is enough to supply about a quarter of current marketed energy consumption. You will note that this 0.017 TW is insignificant compared to the 4.2 TW of solar resource. This is true almost everywhere, not just in the Netherlands; MacKay cites a worldwide estimate http://withouthotair.com/c30/page_235.shtml http://withouthotair.com/c30/page_235.shtml of some 53000 TWh/year (6 TW) of wind resource, which you will note is comparable to the solar resource of just the Netherlands. Current Dutch wind farms total some 6 TW of nameplate capacity and are growing about 10% per year. These produce about 2 TW of power on average, giving about a 30%–40% capacity factor, a bit low for wind but not unreasonable. This represents 2% of the Netherlands' current energy demand, potentially growing to 4% within a decade. https://globalwindatlas.info/area/Netherlands https://globalwindatlas.info/area/Netherlands gives much higher numbers: 500 W/m² at a height of 100 m. This is lower than the solar resource only by a factor of 2, rather than the factor of 400 from MacKay. To resolve the discrepancy, let's look at https://en.wikipedia.org/wiki/Windpark_Noordoostpolder https://en.wikipedia.org/wiki/Windpark_Noordoostpolder, in an area the Global Wind Atlas rates as having 600 W/m² resource. It has a nameplate capacity of 429 MW and generates 1.4 TWh/year, which is 160 MW, a 37% capacity factor. It occupies 8 km², thus giving 20 W/m² on average, halfway between the two numbers. This is a bit better than what you would get with current mainstream solar panels (200 W/m² with a 10% capacity factor gives you 20 W/m² of panel, but angled at 53° you get 12 W per m² of land). But presumably if you blanket a large percentage of the country with windmills, the wind speed will go down. Tidal power is about two orders of magnitude smaller than wind power; MacKay http://withouthotair.com/c30/page_237.shtml http://withouthotair.com/c30/page_237.shtml cites "Kowalik (2004)" as estimating a worldwide practically-extractable resource of 40–80 GW. The Netherlands, with its mere 1900 km of coastline, can't have more than 1% of that total, since it has about 0.1% of the world's total coastline of 1.6 million km. Norwegian hydroelectric energy is almost tapped out, and certainly cannot provide an order of magnitude more energy than at present; there just isn't enough rainfall at high elevations. Moreover, it is a particularly expensive form of energy, not just in flooded upstream lands and the risk of dam breaches but also in construction. Historically it has been competitive with, even a bit cheaper than, fossil-fuel power plants; but it has not experienced the dramatic collapse in pricing that photovoltaic has over the last decade. — ⁂ — Historically, geothermal power has been limited to areas with subterranean water in near contact with magma near the surface. But modern hydrofracking approaches ("enhanced geothermal systems", also more straightforwardly called "hot dry rock") have demonstrated the ability to extract thermal energy from much deeper rock without the necessity for it to already be wet; almost anywhere in the world can do this, using drilling equipment similar to an oil well. An enormous amount of fossil energy is locked in the crust from the last several billion years of radioactive decay. For uses like electricity and transport, though, the trouble is that turning the heat into a useful form requires running that heat through the same kind of steam turbines that are failing to compete with PV in existing coal and oil plants. The resource is enormous, orders of magnitude larger than solar, but even if drilling were free, solar will be cheaper until we start running out of land. Unless heat engines suddenly get a lot cheaper. The geothermal gradient is typically about 30°/km, so typically you need to drill 3 km to even reach boiling, and almost 12 km to reach water's critical point. Most of the heat is "fossil heat"; heat flow out of the earth is currently about 44 TW, but this has been going on for 4 billion years, and most of that heat is still there, several million yottajoules.

kragen 6y ago

Maybe the problem of transporting the energy produced there to China or the Netherlands or something? The problem with selling energy in Morocco or Egypt is that the people there are, by and large, poor, so they can't pay you very much.

jillesvangurp 6y ago
Easier to generate it locally, which people in both China and the Netherlands do.
- kragen 6y ago
  Indeed, and China has the Gobi, but the Netherlands' total solar resource is only a few terawatts, while hundreds to tens of thousands of terawatts are potentially available to be imported. https://www.en-tran-ce.org/custom/uploads/2019/02/Renewable-Energy-January-2019.pdf https://www.en-tran-ce.org/custom/uploads/2019/02/Renewable-... seems to say that last January the Netherlands had 4300 MW(p) of solar PV generating capacity, which produced 0.4 petajoules. If that's correct, the capacity factor of solar PV in the Netherlands averaged 3.5% that month, which is shockingly low, about three times lower than even Germany and nearly ten times worse than equatorial desert solar power plants — although January is one of the worst months, and eyeballing the graphs it looks like the year-round average is closer to 1.3 PJ per month, giving a deplorable 10% capacity factor similar to Germany's. Until and unless panel prices drop dramatically again, situating the panels somewhere sunnier will produce three times as much energy, enough to pay for constructing quite a bit of energy transmission infrastructure. But what happens if panel prices do drop dramatically? Well, if this appalling 10% is a typical capacity factor for utility-scale PV in the Netherlands, the total solar resource is only 4.2 terawatts, of which about 900 gigawatts is available with typical 21%-efficient solar panels. This is an order of magnitude higher than the country's current energy demand, but of course that means that you'd have to cover about 10% of the countryside with solar panels to fulfill that demand from PV alone — and seasonal variability means you'd have to triple that if demand is constant year-round. Covering a third of the Netherlands with PV panels is probably not possible, and covering more than 100% of it is definitely not possible. So the new kinds of energy-intensive industry that will become possible if PV panels get cheaper still will be out of reach of indigenous PV production.
  - jillesvangurp 6y ago
    These back of the envelope calculations are nice. But the underlying assumption of having just one source of energy to cover all needs is wrong. In the Netherlands we have lots of untapped potential in the form of wind, tidal, geothermal and imported hydro-electrical (from Norway). There are only 2 coal plants left (a lot of them were shut down in recent years) and quite a few gas plants and 2 relatively small nuclear plants that are very old at this point (i.e. likely to be shut down at some point). Between all of those sources, I suspect that importing solar from north Africa or China is not going to be attractive from a cost or efficiency point of view any time soon. Offshore wind is getting a lot of investment and dropping panel prices are going to cause a lot of homeowners to put these on their roofs. Lots of people put panels on their roofs already, which in most cases does not fully cover their needs but lowers their energy cost. I wonder if that is included in the report. Reports like this tend to focus on grid providers. A lot of solar generation is happening off grid. Home owners, businesses, EV charging stations, etc. are all use cases where solar panels are commonly used to lower energy cost.
    - kragen 6y ago
      As happened last time I conversed with you, you seem to have reinterpreted my words to mean something that is obviously wrong so that you can disagree with it, and then downvoted my original comment when I had the temerity to rebut you — this despite the fact that my comment explicitly disclaimed the absurd interpretation you imposed on it. — ⁂ — I didn't make an underlying assumption about having just one source of energy to cover all needs. I made an extensively-sourced, well-substantiated calculation that, out of all the sources of energy available in the Netherlands, two of them represent a large enough resource to support an order-of-magnitude increase in energy usage — enhanced geothermal and solar — and neither of these will be economically competitive with importing solar energy from elsewhere. The absurd misinterpretation that underlies your comment is that I'm talking about what will happen in, say, 2021, 2022, 2024, or 2028. But it should be obvious that the Netherlands is not on a path to use an order of magnitude more energy in 2028, nor will solar power in Morocco and the like usher in "new kinds of energy-intensive industry" that quickly. So, as should have been obvious from my initial comment, I'm talking about a longer timescale than that; I think 2040 to 2060. You chose to ignore this and instead interpret my comment as patent absurdity. — ⁂ — Let's talk a little bit about this "lots of untapped potential" and "a lot of investment" and "a lot of homeowners" who "put panels on their roofs" you tout in your comment. — ⁂ — First, what about the homeowners? Well, putting panels on your roof doesn't improve your capacity factor relative to the utility-scale plants surveyed in the reports I cited; in most cases it worsens it, because it's too expensive to reorient the roof to the optimal angle, and most people just use whatever the roof's existing angle is instead of building an extra truss to support the panel. So a square-meter 200-watt (peak) solar panel on a Dutch roof is still going to generate only about 20 watts on average, maybe 15 watts if your roof is at an unfortunate angle. Moreover, there isn't nearly enough roof for rooftop solar to meet a significant fraction of demand. Current marketed energy consumption in the Netherlands is about 60 gigawatts electric, 100 gigawatts total, about 5.6 kilowatts per capita — about 280 square meters of panels per person. Nevertheless, the total solar resource in the country is, as I said, 4.2 terawatts, of which 900 gigawatts would be available with current mainstream solar panels. It should be apparent that Dutch residential rooftop solar can only ever be an insignificantly tiny fraction of even the total current demand. It is an irrelevant distraction. — ⁂ — Of the other sources you mention, three — wind, tidal, and hydroelectric energy imported from Norway — are also insignificantly small compared to the country's solar resource. The fourth, geothermal, represents a larger energy resource than the solar resource, but because it is more expensive to access, it will remain untapped until after the Netherlands starts to experience a scarcity of places to erect solar panels. David MacKay's Europe-wide estimate of the wind power resource per person is 0.4 W/m² http://withouthotair.com/c30/page_232.shtml http://withouthotair.com/c30/page_232.shtml, which works out to 17 GW in the Netherlands' 42865 km². This is enough to supply about a quarter of current marketed energy consumption. You will note that this 0.017 TW is insignificant compared to the 4.2 TW of solar resource. This is true almost everywhere, not just in the Netherlands; MacKay cites a worldwide estimate http://withouthotair.com/c30/page_235.shtml http://withouthotair.com/c30/page_235.shtml of some 53000 TWh/year (6 TW) of wind resource, which you will note is comparable to the solar resource of just the Netherlands. Current Dutch wind farms total some 6 TW of nameplate capacity and are growing about 10% per year. These produce about 2 TW of power on average, giving about a 30%–40% capacity factor, a bit low for wind but not unreasonable. This represents 2% of the Netherlands' current energy demand, potentially growing to 4% within a decade. https://globalwindatlas.info/area/Netherlands https://globalwindatlas.info/area/Netherlands gives much higher numbers: 500 W/m² at a height of 100 m. This is lower than the solar resource only by a factor of 2, rather than the factor of 400 from MacKay. To resolve the discrepancy, let's look at https://en.wikipedia.org/wiki/Windpark_Noordoostpolder https://en.wikipedia.org/wiki/Windpark_Noordoostpolder, in an area the Global Wind Atlas rates as having 600 W/m² resource. It has a nameplate capacity of 429 MW and generates 1.4 TWh/year, which is 160 MW, a 37% capacity factor. It occupies 8 km², thus giving 20 W/m² on average, halfway between the two numbers. This is a bit better than what you would get with current mainstream solar panels (200 W/m² with a 10% capacity factor gives you 20 W/m² of panel, but angled at 53° you get 12 W per m² of land). But presumably if you blanket a large percentage of the country with windmills, the wind speed will go down. Tidal power is about two orders of magnitude smaller than wind power; MacKay http://withouthotair.com/c30/page_237.shtml http://withouthotair.com/c30/page_237.shtml cites "Kowalik (2004)" as estimating a worldwide practically-extractable resource of 40–80 GW. The Netherlands, with its mere 1900 km of coastline, can't have more than 1% of that total, since it has about 0.1% of the world's total coastline of 1.6 million km. Norwegian hydroelectric energy is almost tapped out, and certainly cannot provide an order of magnitude more energy than at present; there just isn't enough rainfall at high elevations. Moreover, it is a particularly expensive form of energy, not just in flooded upstream lands and the risk of dam breaches but also in construction. Historically it has been competitive with, even a bit cheaper than, fossil-fuel power plants; but it has not experienced the dramatic collapse in pricing that photovoltaic has over the last decade. — ⁂ — Historically, geothermal power has been limited to areas with subterranean water in near contact with magma near the surface. But modern hydrofracking approaches ("enhanced geothermal systems", also more straightforwardly called "hot dry rock") have demonstrated the ability to extract thermal energy from much deeper rock without the necessity for it to already be wet; almost anywhere in the world can do this, using drilling equipment similar to an oil well. An enormous amount of fossil energy is locked in the crust from the last several billion years of radioactive decay. For uses like electricity and transport, though, the trouble is that turning the heat into a useful form requires running that heat through the same kind of steam turbines that are failing to compete with PV in existing coal and oil plants. The resource is enormous, orders of magnitude larger than solar, but even if drilling were free, solar will be cheaper until we start running out of land. Unless heat engines suddenly get a lot cheaper. The geothermal gradient is typically about 30°/km, so typically you need to drill 3 km to even reach boiling, and almost 12 km to reach water's critical point. Most of the heat is "fossil heat"; heat flow out of the earth is currently about 44 TW, but this has been going on for 4 billion years, and most of that heat is still there, several million yottajoules.