Climate Change Challenges

The debate about climate change is strangely controversial. Most of the debates seems to focus on the validity of models. I don’t get it.

Obviously, the climate models are incorrect. But that’s not a failure of science, but an inherent property of reality: We are really bad at predicting the future.

All models are wrong; some models are useful. Geroge Box

Models about complex systems – and “Oh my!” is the climate system complex! – are not meant to accurately predict the future. They are created using the available data and scientific hypotheses to create scenarios, bounds of the possible, best guesses about outcomes given certain assumptions. They are meant to make risks of behavior/interventions transparent.

What do we know about the risks of climate change?

A few remarks about models

Let’s distinguish between 3 layers of climate models.

The first level is the basic science: solar radiation, thermodynamics, fluid dynamics, etc.
Most of it is testable at the lab scale. We got a handle on that. I have a lot of trust in this physical basis of the models.

The second level is the system dynamics. The integration of a lot of these well-understood parts into a whole computational system, appropriate discretizations, defining subsystem boundaries, boundary conditions and initial states. Given my own experience with non-linear dynamic systems, this is more of an art than a science. The problem is that you can hardly run experiments on isolate parts of the whole system. It’s hard to only observe the formation of clouds without taking a lot of other variables and (sub)systems into account.
I have a whole lot less trust in these models.

The third level are the economic models built on top of the results of the second stage. They incorporate loss of real estate value, harvests, cost for mitigation etc.
I don’t believe we have a good idea about the economy of any 80 year period in the future. We just don’t know what new technologies will come along or what policies will influence the economy. I don’t trust models on this level.

When we try to gauge the risks of the consequences of human activity on the climate system, using ordinary measures like “cost”, which relies on the third, most uncertain layer of climate models, seems like a rather risky strategy. What is the cost of lost coral reefs? Lost lives? What’s the price of oil in 2084? What will an acre of land in Arizona cost in 2051? What will the regulations on drought-resistant weed be in Pakistan in 2043 How will this affect food security? How do we price in negative Black Swans like climate tipping points? I have no idea. And I’d be surprised to learn that there have been randomized controlled experiments on this.

So I think I will look at the problem from an lower level. What is the root cause of climate change? Anthropocentric climate change is the hypothesis that a rise in the atmospheric levels of green house gases by human activity results in a rise in average temperatures. The main greenhouse gas that’s emitted by human activity is CO2, methane is a distant second. If we want to make sure we dodge the risks of warming, we need to reign in CO2 emissions.
A rise in temperature leads to changes in the weather patterns, which will put existential pressure on a lot of ecosystems. The rate of extinctions is already alarmingly high.
Ultimately, our agricultural systems could fail from droughts and heat stress, people could die in increasingly intense extreme weathers, coastal cities could be threatened by a rise in sea levels and ocean acidification could eradicate the largely unknown ecosystems of the ocean. Scientist have also identified so called tipping points in the our climate system. Irreversible events, like the thawing of permafrost soils or the destabilization of methane hydrates in the ocean, which would release ever larger amounts of methane to the atmosphere and lead to “run away warming”.
The consequences are huge, but we are rather unclear about the odds.
So risks seem to be ill-defined.

What do we need to do to mitigate these risks?

Emissions and CO2 levels

News reports are often looking at the annual CO2 emissions, when they write about climate change. And politicians make bold statements about becoming “climate neutral” by some far away future date. But that doesn’t tell us the whole story.
If we look at the hypothetical scenarios for future carbon emissions, the red scenario looks just horrible, right? “Unlimited growth cant’t work on a finite planet!” So the zero-growth scenario is depicted in yellow. We don’t emit more CO2, so we will halt the problem, right?

But that’s just not true. The average global temperature is influence by the concentration of CO2 in the atmosphere. This means we have to look at the sum of all emissions. Here is what it looks like for the scenarios:

Going to “zero-growth”, the yellow line, doesn’t fare significantly better than the red growth scenario. So that doesn’t solve our problem. What about reducing our emissions to zero? Like the green scenario here:

Wouldn’t that solve our problem? Again, let’s look at the sum of all emitted CO2:

Even the green scenario, which means getting all emissions to zero, emits approximately 50% of the red scenario. So, we’d add about half the temperature increase, we’d expect in the worse scenario.
This means the even the green scenario does not “solve” climate change. It only makes it a little bit less bad. What would a solution look like? Maybe something like this:

Are we near a scenario like that? NOT.EVEN.CLOSE!

As you can see, we’d need to get to negative emissions. We’d need to start to actively remove CO2 from the atmosphere!

There is no solution to climate change that does not involve negative emissions by 2050. Either we find a technology to economically remove CO2 from the atmosphere or we won’t even come close to solving climate change.

What is the cost of removing a tonne CO2 from the atmosphere? Currently, more than $100. So what is the cost of getting the excess CO2 out of the atmosphere? Somewhere around $90-220 trillion. That’s after we reached zero emissions in all sectors of the economy in world of growing energy demand! But that’s the third kind of model. Who knows, maybe it will cost only $5/tCO2 in the future?

The cost estimates assumes that we can build out technologies that are currently in a very early stage of development. Of course, we’d also need huge amounts of energy to run these systems that we currently don’t even have.

That sounds like pretty hard and expensive, right? So let us not be too ambitious. Maybe we can start with going “carbon neutral”? It’s what politicians promise, anyway. Let’s go the “green” route. “Let’s just use solar and wind!”

Renewable Energy

While being popular in certain Western countries, they are still making up only a tiny percentage of total energy consumption globally. This does not mean, that they can’t play a larger role in the future, of course. How likely is this to happen in the relevant timeframe?

To answer that, please consider this: 2/3 of current emissions comes from developing countries. Raising the standard of living means consuming more. At least for a few billion of people in still developing countries. The gap between the average Western European or North American citizen and citizens of Asian, African or South American countries is staggering. They have an awful lot of catching up to do.

Do you think they will use solar for that predicted growth in consumption? It’s cheap right? You hear it all the time in the news! Well…

Solar energy has a range of beneficial attributes: you can test new developments small and scale out rapidly; we have have few new technologies in the pipeline, like “perovskite” solar cells, which utilize a certain crystal structure to archive better material properties. A major advantage is the ability to utilize these types of cells in a new production process, where the cells are printed roll-to-roll in a continuous process, which should be faster and cheaper than the current batch technology. So, long-term, this kind of technology will likely make solar cells available at extremely low prices.
But the main disadvantage still remains: cost.
Don’t get me wrong, all available data indicates that solar energy is already price competitive with fossil fuels, when it generates electricity. But the devil is in the detail. Because you are not really sure, when it will be producing energy.
Wind energy systems have similar problems. Of course they don’t have as pronounces intra-day-cycles as solar systems, i.e. delivering no electricity at night, but they are also suffering from natural seasonal cycles. Additionally, it doesn’t look like there are huge cost reductions for this technology in the pipeline. Wind energy systems are quite mature. (But maybe something like high-altitude wind systems here or here can change the equation)

Today, solar is mostly integrated in grids with preexisting fossil fuel power plants that can flexibly make up the difference between renewable energy production and actual demand. The more solar there is in a given system, the less those flexible power plants will be used. Great! Less CO2, right?

Spoiler: System Costs

Yeah, but you have to build, operate and maintain these power plants. And somebody has to pay for it. In developed countries, you could mandate a higher usage of renewable energy. The lower the utilization of those backup power plants, which translates to higher costs, could be transferred to rate payers and large utility companies.
That’s not possible in a lot of developing countries. Higher electricity prices translates to throwing people back into the poverty they have just escaped from.

So from a system‘s perspective, the price of each additional unit of solar energy will tend to increase beyond a certain point. There are more and more studies coming out that point to this problem.

If you have to build the fossil power plants anyway to get to reliable power generation, why wouldn’t you utilize them? Once built, fossil fuel plants are competitive after all.

“What about storage?” If we make breakthroughs in storage technology, the need to build “backup” power plants will cease. You could “just” build enough storage for your intermittent sources and would not need any fossil fuel power plants. But we do not currently have anywhere near the storage capacities or even a credible plan to scale existing storage technologies to enable seasonal energy storage.
It just doesn’t even look like we should hold our breath for storage for as long as a week. Even enabling short-term storage for electric energy is hard. (Although there are interesting projects like EnergyVault, flow batteries, mountain gravity energy storage, Quidnet ).

So the bottom line is: if you need reliable energy on the cheap right now, you still pick some form of fossil fuels. That’s why India, Indonesia and China are aggressively deploying coal power plants. In 2019, there were 574GW of coal power plants being developed.
Do you think Indian politicians would not love to deploy clean energy? Given the monstrous levels of air pollution in India, I am pretty sure they’d love to. But the energy needed to get out of the most abject poverty is considered more important than the environment at the moment. And there’s not penalty to putting CO2 into the atmosphere.
How likely do you think they will decommission these new fossil power plant capacities, once they are build?

To sum it up: developing nations are expanding their fossil fuel usage, developed countries struggle to stabilize theirs and we can’t rely (yet) on renewable energy and storage alone. To answer the initial question: It doesn’t seem too likely that solar and wind will decarbonize the electricity sector in developing nations in the timeframe it needs to happen.

The larger picture

The catch: electricity only accounts for one quarter of overall CO2 emissions!
Transportation, heating, farming, steel, glass and cement make up the lion’s share of the remaining three quarters. You can’t easily use renewable energy for those sectors. So even 100% renewables on the grid will only get you a quarter of the way. The easiest quarter!

You have seen electric cars, right? I bet you are also familiar with the debate about their range. People are afraid to have a too small battery to get where they want.
Have you seen battery powered trucks yet? No? It’s the same debate. If you have to haul a lot of heavy stuff, you need a lot of energy. That means you need a large battery. Which is also heavy. If you want to go from coast-to-coast, that is a problem. There is a huge trade-off between transport capability and rage.
The same principle applies to ships. And planes. Fossil fuels pack a lot of energy for their weight and volume. For long ranges, current batteries are prohibitively large and heavy.

In a lot of industrial applications you need a lot of heat at high temperatures. You know of course, that burning fossil fuels produces such high temperatures, that’s how we make electricity from them, after all! But just picture a power plant in your head. There is a lot of stuff in there, to convert the heat to electricity. The boiler, all the pipes for steam, the turbines, generators, cooling towers, air filters etc. A good coal fired power plant will convert something like 40% of the energy in the heat to electricity. You throw away three fifth of your heat! The price of the electricity is arguably still competitive with electricity from renewable energies.
When you use fossil fuels for heat only, like in one of those gigantic blast furnaces for steel production, you don’t need all the high-tech, high-cost equipment of a power plant and you don’t throw away 60% of the heat.
If you want to make a dent in the usage of fossil fuels in sectors like steel, you will need to find a way to make carbon-free electricity competitive to heat directly from fossil fuels.

So we will probably not get to zero-emissions with solar and wind and hardly even tackle the other 75% of the emissions with that. Aren’t 80/20 solutions supposed be the other way around!?

But it gets worse: If have written about the economics of climate change previously. To recap there are 3 economic problems: there is no global authority to manage global common goods, multi-party agreements suffer from the prisoner’s dilemma and the cost are invisible to individual economic actors. The Prisoner’s dilemma means that each and every country is better off by all other countries reducing their emissions. But simultaneously, each and every country is even better off, if does not pay for mitigation but all other countries do.
A setting like that incentivizes “selfish” behavior and there is little that can be done to persuade people to adhere to deals, if they can get struck at all.
The Paris Climate Treaty for example does not include any enforcement mechanism and not surprisingly no major country is on course to reach their goals, which were voluntary promises from the beginning.
So there isn’t even an incentive to reduce your emissions.

So maybe we can just consume less?

A lot of people can’t just “consume less”. They are at or below the poverty line. I don’t know what would justify condemning them to stay there forever.
Even if the developed world tried to consume as little as possible tomorrow, there would still be 2/3 of the growing emissions of the developing world and what ever the emissions of all essential sectors, like farming.

I think that realization is what drives a lot of ecologically concerned people to embrace what I call the “trivial solution”: the demand to “drastically reduce the population”, like in Michael Moore‘s latest film “A planet for humans“. I’d call it mass murder, but given that I have never heard how this is to be achieved, I’ll hold my judgement. Maybe there is another solution other than murder to get to the desired population in the relevant timeframe?

If you are not into “reducing the population”, what do we need to do?

What’s the minimum we have to do?

If you want to tackle climate change, you need to have your first negative emissions by 2050. How would we ever get there? Solomon Goldstein-Rose laid out a plan in “The 100% Solution” to get the world to that point. It has five pillars:

1. “Deploy clean electricity generation

2. Electrify equipment that can be electrified

3. Create synthetic fuels for equipment that can’t be or isn’t electrified by 2050

4. Implement various non-energy shifts.

5. Get to negative emissions using sequestration.”

These five pillars are the physical minimum needed to get to a point at which you can start to remove CO2 from the atmosphere. It’s still a far way to solving climate change from there, but it’s a start. No matter what combination of policies, initiatives, public or private action. That needs to happen, if you want to have a shot at the problem.

What does “Deploy clean electricity generation” mean?
We need to build four to six times our current global electricity system from scratch in about thirty years. Using only carbon-free sources.

I can’t overemphasize how unfathomable huge a task this is.

Goldstein-Rose’s estimate of 100-150 PWh/a by the year 2050 translates to 11.5-17.1 TW of average power. . Given a capacity factor for solar of 10-25%, that’s anywhere between 46-171 TW of installed capacity. That’s in the ballpark of Mark Z. Jacobson‘s “100% Clean and Renewable Wind, Water, and Sunlight (WWS) All-Sector Energy Roadmaps for 139 Countries of the World.” It’s the most concrete “plan” for getting to 100% renewable, I have seen. He assumes 11.8 TW of average electricity capacity and estimates 49.9 TW of installed capacity. Which is consistent with the lower bound of Goldstein-Rose’s estimate.
In 2019, there were 115GW of newly installed solar capacity. So we’d need 15-50x the current rate of installation for 30 years starting now. Looking at wind power, there were 60.4GW of wind power installed in 2019. So at a capacity factor of 25% we’d need 25x that for 30 years starting now. We could of course also have a mixture of those technologies. And that also means 25-50x the land use, which adds up to quite astonishingly large areas.

Oh, and we would need hundreds to thousands of times our current storage capacities.

Jacobson estimates a cost of $115 trillion for the capital and a electricity costs of $0.11/kWh. He arrives at that cost by assuming $0.008/kWh in storage costs.
I have no idea how we would even begin to store electricity seasonally for $0.10/kWh, let alone for $0.008/kWh. On the other hand, there are super optimistic forecasts of around $0.01/kWh for solar, so maybe $0.11/kWh is not that far off after all.
He also assumes that 59-85% of all energy demand is flexible. I translate that to significant changes in life-styles. I don’t know, if that’s feasible. And I don’t know if that cost of electricity is bearable for developing nations.
But I do know that it’s not competitive without a price on carbon or pricing in other externalities, like health issues due to particle emissions. Something that’s currently not happening.

Even with optimistic assumptions, I’d like to side with Goldstein-Rose:”We don’t have ‘all the technology we need’ affordable enough even in countries where political will could mandate slight cost increases.”
And there are large parts of the economy for which full decarbonization means massive increases in costs.

We have no option but to buckle up and innovate our way out of it!

I don’t believe there will be anything near to a solution in a frame of “scarcity”, where all countries and all citizens have to do “their part” and “sacrifice” as much as possible for the sake of all other people. I might be a pessimist, but I just don’t see that happening.
What I think will work: making it profitable to work on climate.

We have to shift our frame of reference from: “How can I sacrifice the least on this and make the suckers pay?” to “How I get the fattest slice of the money out of this?”

Develop a source of dispatchable energy that can be deployed in India for 0.03$/kWh and you will never have to talk about “ending coal” ever again. It will happen there automatically.

The US natural gas market is a good proxy for it. The shale gas revolution was brought about by new technologies in the drilling industry and made natural gas so cheap that new coal power capacities are rarely economical.
There was no law necessary to diminish coal’s role in energy production, to the contrary, even the active promotion of coal by US President Donald Trump could hardly make a dent in the decline of coal for new power generation facilities.
That’s the power of economics at work!

Leading the development and deployment of low-carbon technologies has to be made profitable. There is no need for international agreements, if it is profitable to adopt new technologies. Every single country could individually change our current path, if it creates the necessary innovation.
The single most helpful policy change would be a price on carbon. I have written previously about how a few countries could implement a global carbon price. But it’s not strictly necessary. The second most helpful policy would be a massive ramp up in investments in R&D.

Do we have anything else left in the tank?

The nuclear option

At a capacity factor of 93% nuclear power plants are highly reliable. The middle of Goldstein-Rose’s estimate, 125 PWh by 2050, translates to roughly 12500 large nuclear reactors (1250MWe). Without the need for energy storage. How much would that cost?

If we take the $25 billion it cost to build 4 South Korean APR1400 units in the UAE, that would translate to $78.125 trillion. They have an expected lifetime of 60 years, instead of 20 years for most renewable energy infrastructure.

According to a rather optimistic estimate and counter-proposal to Jacobson’s plan, the so called Liquid Fission Road Map, going nuclear with a new type of nuclear power plant would cost an estimated $15 trillion in capital costs and produce electricity at $0.03/kWh.

We don’t have all of the technology yet, but we are pretty sure we know how to get there.
Wright’s law will probably be a major driver in this regard, IF – and that’s intentionally big – regulators take a sensible approach to the risks for nuclear power plants. Which, I hope, will lead to massive private investments. Are $100 trillion in saved capital and large, virtually guaranteed profits from running the backbone of civilization ever after that a fat enough pie to work towards?

I will look at some of the options and challenges for nuclear energy production in the future. It’s a fascinating topic!

Summary

There has never been a greater industrial challenge posed to mankind. The scale of it is just mind-boggling!
I don’t think that a lot of so-called climate activists have a grasp on even the magnitude of what a “solution” to climate change entails.
Claiming it’s only the political will that’s missing is almost disqualifyingly wrong – we are lacking a whole lot of affordable technology!
Demands of consumption reduction are comically bourgeois and shortsightedly focused on developed nations.
Demands for “population reduction” are borderline genocidal.

Honestly, I can’t imagine that we will meet this challenge. I think it’s way more likely that we will see some combination of efficiency measures, all sorts of clean energy deployment, CCS, mitigation, a little bit of geoengineering and a lot of hoping for the best.
But the more clean energy we deploy, the less we will have to press our luck.

So we will look at what I believe to be our best options. See you soon!