The world's foremost climate scientist, James Hansen, had some blunt things to say about
Hillary Clinton's plan to put half a billion solar panels on rooftops across the country.
"It’s just plain silly. No, you cannot solve the problem without a fundamental change, and that means you have to make the price of fossil fuels honest. Subsidizing solar panels is not going to solve the problem ... We have two political parties, neither one of which is willing to face reality. Conservatives pretend it’s all a hoax, and liberals propose solutions that are non-solutions."
-- Dr. James Hansen, quoted in The Guardian
Yes, I bolded that last for a reason:
he's right. Just like he's always been right.
Every ton of CO2 we emit stays in the air for centuries, and will continue to warm the planet long after we're dead, and continue to harm the climate long after we have reaped the benefits of emitting it. It is simply immoral that poor farmers in Bangladesh will eventually lose everything, including possibly their lives, so that rich kids on Long Island can play Halo 3. But it doesn't have to end that way.
US non-fossil electricity needs, and the Democratic response. Are we embarrassed yet?
About 40% of US primary energy consumption is devoted to electricity. The other 60% is used for transportation, space heating, and industrial process heat. About 70% of our (US) electricity is fossil fueled, and nearly all of the non-electrical energy use is fossil. So we do the math, and 88% of all of our energy is fossil. And we have to get rid of all of that, rapidly.
Hillary's plan references the Deep Decarbonization Pathways Project, a serious attempt by serious people to figure out how that might actually be possible in a timeframe that might actually save the planet. DDPP's goal is an 85% reduction in CO2 emissions (from 2005 levels) by 2050.
DDPP offers several scenarios whereby this might be achieved, but all of them share this general outline: we decarbonize the electric grid first; then, we electrify all the remaining uses of fossil fuels. We use non-fossil electricity for transport or to make synfuels, and replace furnaces with heat pumps. Of course some non-electrical uses can be converted sooner, and should be, but that's the general idea. The upshot is this: to decarbonize the economy, our electrical use will have to go up, not down. Because we're going to be converting most of that other 60% over to non-fossil electricity. And that's going to take a lot of Watts.
How much electricity will we need?
DDPP estimates that our electricity generation will have to nearly double (increase of 60% to 110%). But I only looked that number up after doing my own analysis, which came to the same conclusion: an increase of 82% -- almost exactly the middle of DDPP's range. So in this diary I will stick to my own numbers, and we can expect that they won't be too far from DDPP's.
First, I'm assuming that total primary energy demand will grow only very slowly over the next 35 years. Currently (2014) we use 2299 MTOE (million tonnes of oil equivalent) in primary energy, according to the BP Statistical Review. Based on decelerating growth in the past decades, I project that in 2050 growth will have flattened entirely and we will be using 2374 MTOE, not much different than now. (That is actually an extremely optimistic projection, assuming either the economy continues its long-term slowing trend, or that energy conservation can actually reduce overall demand. I have argued elsewhere that the latter is probably false.)
After juggling the numbers a bit, meeting that goal will require 7829 Terawatt-hours (TWh) of generation, up from our current 4297, and it will all have to be non-fossil.
In 2014 we managed 1371 TWh generated from all non-fossil sources combined. That leaves us 6458 TWh of new annual non-fossil generation we need to add in 35 years, a rate of 186 TWh per year of new non-fossil generation added every single year for the next 35 years.
And Hillary's plan at the end of 2020 manages to get us all the way up to 37 TWh per year of new solar generation added, in its last best year. That's assuming the current national average of 14% capacity factor for solar. (Yes, it's higher in the desert, but Hillary's plan is heavy on rooftops, so the number is reasonable.)
Meanwhile, the new EPA rule proposes to save us the equivalent of just 524 TWh of coal-fired generation, about 8% of what we will need by 2050.
Ouch.
Hillary's plan doesn't address the shortfall (at least not yet, though she has implied there may be more to come); neither does Bernie, nor Martin. The Obama climate agreement with China is no better.
We are slow-walking our way toward planetary suicide, one plan at a time.
Yes We Can!
It's worth noting that a number of countries around the world have already decarbonized their grids, or nearly so. Even some third-world nations, like Zambia, Paraguay, and Nepal, have decarbonized with hydropower, as has Norway. France did it with nuclear. Other places have used combinations of those, like Sweden, Switzerland, and Ontario. And Iceland did it with geothermal. So yes, it is possible. But can it be done that fast?
To compare nations of very different sizes, like Finland and the US, it's better to have a metric that accounts for those changes on a per-capita basis. Using the current US population (319 million), we would need 578 kWh of new non-fossil generation per person per year to decarbonize the economy by 2050, if we start this year. If we wait until 2017, we would need 619 kWh per person per year. If we don't start until 2020, we would need 675 kWh per person per year.
I made a list of the fastest non-fossil electrical build rates every nation has ever done in a year. That list tends to have big spikes when a small nation installs a big nuclear reactor, and since a reactor takes typically 7 years or so to build, I took 7-year averages instead to even out the spikes. I wanted to show a top-ten list, but it ended up being more like a top-35 list, with a few additions.
Just to be clear: we need that 578 kWh per capita of
new fossil-free generation
every single year until 2050. We can knock 47 off that number after the EPA rule is fully implemented in 2030. And if we used the average population of the US from now until 2050, we could knock about 40 off of the US requirements, but it's still going to be a pretty big number no matter how you figure it.
The first thing to notice about that graph is that the top 13 places are all held by nuclear builds; and 19 of the top 25 places are held by nuclear. Solar, everybody's favorite lovechild, is a weakling, way, way down at the bottom -- even in Germany.
The #1 non-fossil build rate, in Sweden, has led Qvist & Brook (2015) to suggest that the entire world could decarbonize electricity in 30 years by following their example.
Hillary's plan is ambitious in one respect: it assumes that we can take the biggest long-term average solar build rate in world history, and increase it by 50% or more. But even that ambitious plan is simply not going to do the job, and it's not even close. We also have to take the biggest wind build rate in world history, and increase that by 50%. And we also have to take the biggest nuclear build rate in US history, and increase that by 50% too.
And we have to do all of those things at the same time. Then we might have a fighting chance to save the climate. Anything less is simply inadequate. And that's going to take a major national commitment of the kind not seen since the Apollo program.
Here's what we need:
- Hansen sensibly advocates a revenue-neutral carbon tax, with the revenue rebated to the population on a per-capita basis. Thus people with the biggest carbon footprint pay more, while those with the smallest carbon footprint end up getting more in rebate than they pay in taxes. For those in the middle, it's a wash.
- Then we gradually jack up the carbon price, until fossil is squeezed out of the market.
- A tariff on goods imported from countries that do not put a price on fossil carbon, that reflects the carbon intensity of the imported goods. Nations exporting to us can then avoid the tariff by putting a price on carbon equal to our own. We're the big dog in the world economy, and it's time to show some teeth.
- A levelized playing field for subsidies of non-fossil electricity. Hillary wants big subsidies for solar, and that's great, let's do it. But let's have the same for wind and nuclear and hydro and geothermal. Everybody who installs non-fossil generation needs to be rewarded, and that's the way to bring Republicans (some of them, anyway) on board too. Nuclear is a heavy lifter, and with many nuclear plants in red states, we can rectify the currently unbalanced situation and bring some GOP votes along with it.
- Finally, once we get our emissions under control, we need to seriously think about methods of geoengineering that might claw back out of the atmosphere some of our past emissions. It's our mess, and we should be the ones to deal with it.
In my view, the best way to decarbonize is the fastest way, and given that resources are limited, the fastest way is also the cheapest way:
* Hydro and geothermal where available.
* Wind and solar up to their curtailment points.
* Nuclear for the rest.
This is roughly the same as the DDPP's high-nuclear scenario (which, in spite of its name, foresees massive increases in US wind and solar production).
The task we face is huge, but it's not impossible. Sweden did it, Finland did it, and France did it, and there is no reason we can't do it too. Let's get to work.
Can we do it without nuclear?
Short answer: no.
Longer answer: We still need to keep the lights on at night when the sun isn't shining, and when the wind isn't blowing. Right now we do that with fossil, which is what takes over on calm nights. So how do we do that without fossil? There are only three choices: storage, transmission from other regions, and nuclear. Hydro and geothermal are too geographically limited for the country as a whole (but are great where available).
Capacity factors for wind and solar are so low that if we go to battery backup of renewables, we would need in the ideal case 3 to 4 GWh of storage for every GWh of production. In other words, roughly 25% of our electricity would be coming directly from renewable generation, and the remaining 75% (roughly) would be coming from stored renewable energy. In reality, even a 3-to-1 ratio is not enough, because we would need to allow for some times when we could go many days in a row without Sun or wind. Reliability demands would increase that to about 5-to-1.
Since both wind and solar have strong daily variability, a 1-day charge cycle time would be typical on a renewable-plus-battery grid. Our 2050 electric grid will be generating 7829 TWh a year, which means that we would need about 21 TWh every day. With a 5-to-1 ratio to allow for bad days (and even bad weeks) we would need about 17 TWh of batteries on the grid. Current Li-ion batteries are getting about 1000 charge cycles, but let's optimistically assume that we could get 5000 charge cycles before replacement, which means we would have to replace the batteries once every 14 years or so -- or 1/14 of the total every year. That's 1.2 TWh per year we would have to replace. And the current world production of batteries is .035 TWh per year. In other words, world production of batteries would have to increase by 35 times just to support the US electric grid, and that's not counting the fact that we would be replacing gasoline cars with EVs at the same time.
Not happening.
But what about transmission? Renewable energy is best transmitted in direct current, via high-voltage DC (HVDC) lines. The most recently completed HVDC line in North America, the West Alberta line, began operating in March of this year, and it came in at $6 million per mile for a 1 GW line. And that's over sparsely populated terrain where rights of way are cheap. The Northern Pass HVDC line, planned in New Hampshire at 1.2 GW, is expected to cost $7.5 million per mile, a similar number.
At that price, it's cheaper to build a nuclear plant next door than to transmit wind power 400 miles, even if there aren't any wind turbines at the end of the line. Let's do the math: the new nuclear reactors at Vogtle in Georgia just made headlines because the latest cost overrun pushed the cost to $7.5 billion per reactor. Let's assume that it's going to be $8 billion by the time it's actually done. Each reactor is 1.117 GW, so that's $7.16 billion per GW. Nuclear reactors have a capacity factor of 90%, meaning they run 90% of the time. The capacity factor of wind varies with location, but on average it's about 30%, so we would need about 3 wind farms of 1 GW each to replace the nuclear plant. Which means we need three HVDC powerlines at $6 million per mile, or a total of $18 million per mile for the same 1 GW. So we divide the $7.16 billion nuclear cost by $18 million per mile powerline cost, and the breakeven point is 398 miles -- without the cost of the wind farms.
If you add the cost of the wind farms, the breakeven point is just 100 miles. And at that point, you have eliminated the reason for building the powerline in the first place, which was to make sure you would have some windfarm somewhere that was supplying you power when the wind died locally. Because 100 miles isn't far enough away to be out of your own local weather.
And the nuclear plant is safer, more reliable, and lasts three times longer.
This diary is Part VI of GETTING TO ZERO: Our non-fossil energy future.
Part I of GETTING TO ZERO: The size of the problem.
Part II of GETTING TO ZERO: Is renewable energy economically viable?
Part III of GETTING TO ZERO: Why energy efficiency will not save us.
Part IV of GETTING TO ZERO: The hidden CO2 emissions from renewables.
Part V of GETTING TO ZERO: Deep Decarbonization report: high-renewable path costs 4x more than high-nuclear alternative