Guest Post by Willis Eschenbach
After reading some information at Friends of Science, I got to thinking about how impossible it will be for us to do what so many people are demanding that we do. This is to go to zero CO2 emissions by 2050 by getting off of fossil fuels.
So let’s take a look at the size of the problem. People generally have little idea just how much energy we get from fossil fuels. Figure 1 shows the global annual total and fossil energy consumption from 1880 to 2019, and extensions of both trends to the year 2050. I note that my rough estimate of 2050 total annual energy consumption (241 petawatt-hrs/year) is quite close to the World Energy Council’s business-as-usual 2050 estimate of 244 PWhr/yr.
Figure 1. Primary energy consumption, 1880-2019 and extrapolation to 2050. A “petawatt-hour” is 1015 watt-hours
So if we are going to zero emissions by 2050, we will need to replace about 193 petawatt-hours (1015 watt-hours) of fossil fuel energy per year. Since there are 8,766 hours in a year, we need to build and install about 193 PWhrs/year divided by 8766 hrs/year ≈ 22 terawatts (TW, or 1012 watts) of energy generating capacity. (In passing, for all of these unit conversions let me recommend the marvelous website called “Unit Juggler“.)
Starting from today, January 25, 2021, there are 10,568 days until January 1, 2050. So we need to install, test, commission, and add to the grid about 22 TW / 10568 days ≈ adding 2.1 gigawatts (GW, or 109 watts) of generating capacity each and every day from now until 2050.
We can do that in a couple of ways. We could go all nuclear. In that case, we’d need to build, commission, and bring on-line a brand-new 2.1 GW nuclear power plant every single day from now until 2050. Easy, right? …
Don’t like nukes? Well, we could use wind power. Now, the wind doesn’t blow all the time. Typical wind “capacity factor”, the percentage of actual energy generated compared to the nameplate capacity, is about 35%. So we’d have to build, install, commission and bring online just under 3,000 medium-sized (2 megawatt, MW = 106 watts) wind turbines every single day from now until 2050. No problemo, right? …
Don’t like wind? Well, we could use solar. Per the NREL, actual delivery from grid-scale solar panel installations on a 24/7/365 basis is on the order of 8.3 watts per square metre depending on location. So we’d have to cover ≈ 100 square miles (250 square kilometres) with solar panels, wire them up, test them, and connect them to the grid every single day from now until 2050. Child’s play, right? …
Of course, if we go with wind or solar, they are highly intermittent sources. So we’d still need somewhere between 50% – 90% of the total generating capacity in nuclear, for the all-too-frequent times when the sun isn’t shining and the wind isn’t blowing.
Finally, an update. A well-informed commenter below says:
I think you missed something, Willis
That 22 TW is average power. But generating plants, transmission facilities, transformers, circuit interrupters, and all that stuff, must be sized for the PEAK demand.
Most distribution systems in the US have a peak to average (PtA) ratio of around 1.6 to 1.7. Except for the New England ISO which is running around 1.8. Some systems in Australia have an annual PtA ratio of around 2.3. I expect Arizona would run that high taken in isolation, which, of course, it never is.
Take 1.8 as an estimated overall PtA ratio, you need to meet a peak demand of 22 * 1.7 terawatts or 37.4 TW.
But no power system can survive with generation equal to demand. So add 15% for reserves for when parts of the system are down because of maintenance, failures, or the like. The result is, you need peak generation of 43 TW. So roughly double all of your numbers as to what needs to be built.
And guess what? He’s right. We can’t just provide for average demand. We have to provide enough power for the hottest days in the summer, and for the coldest days in the winter. So we need double the numbers I gave above.
However, there’s another factor to consider. This is that there is not as much heat loss in electric generation and electric cars. Using fossil fuels for generation and transportation is less efficient. So we won’t need to replace the full total of fossil fuels.
At present, about 60% of the fossil energy is lost as heat. However, not all of this inefficiency will disappear if we switch to an all-electric world … and some will increase. For example, overall transmission and congestion losses for the electrical grid are on the order of 15%. So if we are powering homes and industry and transportation via electricity, those losses will be greater, not less.
In addition, while electric car running efficiencies are greater than internal combustion engines, the batteries are very energy-intensive to make. And internal combustion engines use waste heat for heating the car interior, while battery cars use electricity. So the efficiency gains there will not be as great as they might seem.
Next, in some sectors there will be no reduction in losses. If the heating of a building is switched from gas to electric, there is no gain in efficiency, because the same amount of heat is still being lost through the walls and the roof.
Overall, due to increases in efficiency, we’re likely to have to replace only about half of the current fossil fuel use with electricity. So that offsets the doubling mentioned above to allow for peak consumption.
To summarize: to get the world to zero emissions by 2050, our options are to build, commission, and bring on-line either:
• One 2.1 gigawatt (GW, 109 watts) nuclear power plant each and every day until 2050, OR
• 3000 two-megawatt (MW, 106 watts) wind turbines each and every day until 2050 plus a 2.1 GW nuclear power plant each and every day until 2050, assuming there’s not one turbine failure for any reason, OR
• 100 square miles (250 square kilometres) of solar panels each and every day until 2050 plus a 2.1 GW nuclear power plant each and every day until 2050, assuming not one of the panels fails or is destroyed by hail or wind.
I sincerely hope that everyone can see that any of those alternatives are not just impossible. They are pie-in-the-sky, flying unicorns, bull-goose looney impossible.
Finally, the US consumes about one-sixth of the total global fossil energy. So for the US to get to zero fossil fuel by 2050, just divide all the above figures by six … and they are still flying unicorns, bull-goose looney impossible.
Math. Don’t leave home without it.
My very best wishes to everyone, stay safe in these parlous times,
PS—As always, to avoid misunderstandings I request that when you comment, you quote the exact words that you are discussing so we can all be clear about who and what you are referring to.
Hard Copies: Someone said they couldn’t get this to print from WUWT. So I selected the whole document from the title to the end and copied it. I pasted it into Microsoft Word. Then I cleaned up the formatting and saved it to my Dropbox, where you can access it here.
I also saved it as a PDF file for those who don’t have Word. It’s here. However, because it’s a PDF, the links to other documents are not active.
Update re $$$: “Top consulting firm McKinsey has calculated that the net-zero emissions targets set by global governments and championed by the United Nations would cost the public a staggering $275 trillion by 2050, or around $30 billion per day for the next 25 years.” Full article here.
Update re Efficiency: Several people have commented that we don’t need to replace all of the energy provided by fossil fuels, since a lot of it is lost as heat and won’t be if we go electric. My calculations indicate that the savings will be nowhere near what they claim, because for many things like home and office heating the losses are not dependent on the methods used to provide the heat. And electric systems have their own losses, such as transmission losses, which will increase if we go all-electric. Finally, solar and wind require 24/7 spinning backup to replace their generation at a moment’s notice … and at present that’s only practical with fossil fuels, and it requires the spinning backup to run at very low efficiency.
But heck, divide all my numbers by 2, and it’s still flying unicorns, bull-goose looney impossible.
Technical Note: These figures are conservative because they do not include the energy required to mine, refine, and transport the necessary materials, plus provide the energy needed to actually build the reactors, wind turbines, or solar panels. This is relatively small per GW of generation for nuclear reactors but is much larger for wind and solar.
They also don’t include the fact that wind turbines have about a 20-year lifespan, so after 20 years we’ll have to double the turbine construction per day. And with solar the lifespan is about 25 years, so for the last five years, we’ll have to double the solar construction per day. And then we will have to decommission and dispose of millions of wind turbines and hundreds of thousands of square miles of solar panels …
The figures also don’t include the fact that if we go to an all-electric economy we will have to completely revamp, extend, and upgrade our existing electrical grid, including all associated equipment like transformers, power lines, circuit interrupters, and switching stations. This will require a huge investment of time, money, and energy. And this extends into the homes, as every home like mine that’s heated by gas and uses gas for water heating and cooking will need to greatly increase the electrical service to the house and install an electric furnace, stove, and water heater.
Finally, since nuclear power plants take about a decade from site selection to hookup, we don’t have until 2050 to start building them. We only have until 2040, about 2/3 of the time. So we’ll need to build ~ 50% more nuclear power plants per day to get there by 2050
So in terms of energy, these are still conservative figures.
They also don’t include the cost. The nuclear plants alone will cost on the order of US$170 trillion at current prices. And wind or solar plus nuclear will be on the order of US300 trillion, plus decommissioning and disposal costs for wind turbines and solar panels.
Finally, the cost of converting all the individual homes, businesses, and buildings around the world that use gas for heating, cooking, and water heating will be enormous. Who will pay for that?
And this doesn’t touch the cost of the land for siting the windmills and the solar panels, which will be stupendous. Here’s some information from California regarding how hard it is to find suitable land for solar power.
… Another issue is the fact that such solar ‘farms’ require huge tracts of land. The Bureau of Land Management (BLM) has been tasked with finding 24 tracts of public land of three square miles each with good solar exposure, favorable slopes, road and transmission line availability. Additionally, the land set aside for utility-scale solar farms must not disturb native wildlife or endangered species such as the desert tortoise, the desert bighorn sheep, and others. The wildlife issue has proved to be a contentious one. Projects in California have been halted due to the threat caused to endangered species resulting in a backlog of 158 commercial projects with which the BLM is currently contending.
Note that the BLM is having trouble finding a mere 75 square miles of land for solar power generation that doesn’t have too much impact on the environment, and we’re talking about building 200 square miles of new solar power per day …
So it is even more impossible … speaking of which, is it possible to be more impossible?
Because if it is possible … this is it.