Global warming can often feel overwhelming, given its political, social, and economic complexities. From a purely engineering perspective, though, it is surprisingly simple. There is a clear goal and a bounded set of technological tools to achieve it — just the kind of problem engineers like to solve.

The clear goal is net-zero global greenhouse gas emissions by 2050, a target around which much of the world is coalescing in the wake of the recent IPCC report.

Reaching global net-zero is necessary to stabilize the atmosphere at any temperature. Otherwise, it continues warming. “The difference between one and a half degrees, two degrees, and two and a half degrees [of warming] is functionally just the amount of time you have to achieve net zero,” says Julio Friedmann, an energy researcher at the Center for Global Energy Policy at Columbia University. Failing to reach net zero means failing to stabilize the atmosphere.

The term “net-zero emissions” means that for every ton of carbon released from the geosphere into the atmosphere (through mining, drilling, and burning of fossil fuels), one ton must be returned from the atmosphere to the geosphere, either through natural means like absorption in oceans, soil, and plants, or through industrial carbon capture and sequestration. Getting to net zero means reducing emissions as much as humanly possible and burying enough carbon to account for emissions that can’t be eliminated.

A chart showing carbon neutral and carbon negative scenarios.
Net zero, and then negative.

Net zero is the clear goal. The tools available for achieving it are clean energy technologies. Given the time it takes for new technologies to scale up to mass-market significance, the 2050 target will almost certainly be met (if at all) with clean energy technologies that currently exist. Some of them may still be in the early stages of development, but they’re already out there somewhere. It’s a large set of tools, but a bounded set.

From an engineering perspective, the central question is whether the tools available are up to the task required of them.

The International Energy Agency (IEA) has recently set out to answer that question, under the rubric of its Energy Technology Perspectives (ETP) program, which this month issued its latest Clean Energy Innovation report.

The (comprehensive and fascinating) report surveys the field of clean energy technology and determines where various technologies are on the development curve and where they must get to achieve net zero by 2050. It reveals a problem that is at once politically daunting and, from an engineering perspective, eminently solvable — even, or perhaps especially, in the Covid-19 era.

The IEA is trying to give new technologies a chance

The IEA has been criticized in the past for being too conservative in its modeling, for underestimating the rapid pace of development in cleantech and thus overestimating the cost of decarbonizing. (I covered the controversy in detail in this post.) The ETP program is in part an effort to respond to those criticisms.

There are four basic stages that new technology goes through in the process of scaling up to mass-market significance. Quoting from the IEA report:

  1. Prototype: A concept is developed into a design, and then into a prototype for a new device (e.g. a furnace that produces steel with pure hydrogen instead of coal).
  2. Demonstration: The first examples of a new technology are introduced at the size of a full-scale commercial unit (e.g., a system that captures CO2 emissions from cement plants).
  3. Early adoption: At this stage, there is still a cost and performance gap with established technologies, which policy attention must address (e.g., electric and hydrogen-powered cars).
  4. Mature: As deployment progresses, the product moves into the mainstream as a common choice for new purchases (e.g., hydropower turbines).

Because technology development is uncertain and difficult to predict, IEA has typically only used technologies from categories 3 and 4 in its models, which is one reason its scenarios have been seen as unduly conservative.

the stages of tech development IEA

The Clean Energy Innovation report responds directly to that concern. It surveys some 400 clean energy technologies in four key areas (electrification, carbon management, hydrogen, and bioenergy), determining their stage of development. And it models what would happen if all the technologies in the early stages were rapidly accelerated to maturity. It calls this new scenario its Faster Innovation Case.

The report goes into considerable detail about the impact of Covid-19 on global energy innovation. I won’t get into the details; the upshot is that the entire world is at an inflection point.

As of now, the coronavirus has hit energy companies hard and slowed innovation spending. Even as the recession produces a short-term dip in emissions, it could delay longer-term decarbonization, which would only make the eventual task more difficult.

But countries across the world are considering trillions of dollars in recovery and stimulus spending to salvage their economies. That presents an enormous opportunity to kick-start the cycles of building and innovation that will be necessary to hit the 2050 target. If they are smart about it, countries — especially wealthier countries like the US — can deliberately accelerate the development of early-stage clean energy technologies and get them on track to do the enormous work that will be asked of them by midcentury.

Many technologies that will be needed for deep decarbonization are nowhere near ready

The IEA begins by determining how ready current clean energy technologies are to meet the UN’s Sustainable Development Scenario (SDS), which would reach global net-zero emissions by 2070 and stabilize global temperature rise at 1.8°C (along with meeting several other sustainable development goals).

In the energy sector, IEA identifies four key approaches to decarbonization that are lagging technologically:

  1. Electrification of end uses, particularly heating and transportation
  2. Carbon capture, utilization, and storage (CCUS)
  3. Low-carbon hydrogen and hydrogen fuels
  4. Bioenergy

Within those four approaches, IEA assesses more than 400 separate technologies. What is remarkable, and disheartening, is how few of them are on track to meet the SDS goals.

The report does a deep dive on the component technologies in each of the four areas. Here’s how it lays out electrification:

IEA, tech readiness in the power sector IEA

As you can see on the left, many zero-carbon power generation technologies are either mature (blue) or in early adoption (green) and scaling up. But electricity infrastructure (center column) is lagging and electrification of heavy industry (middle of the right column) is practically nowhere near.

Nonetheless, electrification is probably the best of the four approaches. Here’s CCUS:

IEA tech readiness in CO2 IEA

That’s a lot of yellow and orange — almost nothing in early adoption or mature, ready for market.

To see similar breakdowns of the other areas, check out the interactive dashboard IEA built to show all 400 technologies.

Altogether, “around 35% of the cumulative CO2 emissions reductions needed to shift to a sustainable path come from technologies currently at the prototype or demonstration phase,” the report says. “A further 40% of the reductions rely on technologies not yet commercially deployed on a mass-market scale.”


As IEA notes, “this calls for urgent efforts to accelerate innovation.”

All those early-stage technologies need to be pushed forward to bring down costs and achieve mass-market adoption. “Without strong and targeted R&D efforts in critical technologies,” IEA says, “net-zero emissions are not achievable.”

Pushing technology into overdrive

If clean energy technology isn’t even ready to hit the SDS’s net-zero-by-2070 target, how can it possibly hit net zero by 2050?

That is the purpose of IEA’s Faster Innovation Case (FIC): It models what would happen if all those early-stage clean energy technologies were rapidly accelerated through the stages of innovation, twice as fast as they are in the SDS.

IEA is clear that the FIC is not a recommendation. It’s a somewhat idealized exercise that would almost certainly be impossible in practice. Broad adoption of new energy technologies can take 80 years or more. Even some of the fastest cycles of adoption in recent history — say, LED light bulbs — took between 10 and 30 years.

innovation cycles IEA

In the FIC, all early-stage clean energy technologies would match that pace. “CO2 savings from technologies currently at the prototype or demonstration stage would be more than 75% higher in 2050 than in the Sustainable Development Scenario,” IEA reports, “and 45% of all emissions savings in 2050 would come from technologies that have not yet reached the market.”

One reason those early-stage technologies could play such an outsize role is that they could help solve the problem of carbon lock-in. A certain amount of future greenhouse gas emissions are “locked in” by current investment commitments in dirty power plants and factories.

Lots more emissions are about to be locked in by the next round of investment. Investment cycles for some energy technologies, especially big industrial equipment, are in the 20- to 25-year range. “If the right technologies in the steel, cement and chemical sectors can reach the market in time for the next 25-year refurbishment cycle – due to start around 2030 – they can prevent nearly 60 gigatonnes of CO2 emissions (GtCO2),” IEA reports.

locked-in emissions IEA

To avoid large-scale carbon lock-in, rapid innovation is crucial.

How governments can accelerate innovation

Clean technology is not currently on track to do what the world will ask of it in coming decades. To get there, all the world’s major governments will need to commit to a concerted effort to accelerate its development.

Here, quoting from the report, are the five recommendations IEA makes to governments to speed things up:

  1. Prioritize, track and adjust. Selecting a portfolio of technologies to support requires processes that are rigorous and flexible and that factor in local needs and advantages.
  2. Raise public R&D and market-led private innovation. Different technologies have differing needs for further support, from more public R&D funding to market incentives.
  3. Address all the links in the value chain. In each application, a technology is only as close to market as the weakest link in its value chain, and uneven progress hinders innovation.
  4. Build enabling infrastructure. Governments can mobilize private finance to address innovation gaps by sharing the risks of network enhancements and demonstrators.
  5. Work globally for regional success. The technology challenges are urgent and global, making a strong case for co-operation which could draw on existing multilateral forums.

These are all pretty self-explanatory, and I won’t get into the report’s extremely deep weeds — I’ll just make a couple of general points about innovation policy.

First, spending more money is crucial, but it’s not enough. Good innovation policy requires a plan, forward-thinking analysis of the technology landscape and regional and local needs, and some regular assessment and self-correction. There’s no substitute for good governance.

Second, innovation does not mean handing money to technologists and waiting to see what they produce. A big part of accelerating innovation is building — building the technologies themselves, to take advantage of learning-by-doing, and building the infrastructure new technologies need to grow. To scale up clean energy technologies in time, building has to start now and continue at a headlong pace for decades. Friedmann summarizes IEA’s conclusions this way: “every week is infrastructure week, for the next 30 years.”

Just consider what’s involved in the Faster Innovation Case. “Robust market deployment of current prototypes would need to start right after the completion of only one single commercial-scale demonstration,” the report says, “which,” it adds with droll understatement, “is not common practice.”

That will mean taking lots of big shots on risky bets, some of which will inevitably fail. Doing so requires political systems and electorates willing to be patient and forgiving. (You will recall that President Barack Obama’s clean energy loan program, which was if anything too conservative, was forever poisoned in the public mind by Republican demagoguing of Solyndra, a company that failed after receiving government loans.)

In the FIC, “demand for hydrogen and hydrogen-based fuels would grow by almost 25% in 2050 over the Sustainable Development Scenario,” the report says, “requiring, for example, almost two new hydrogen-based steel plants (today at prototype stage) to be installed each month from now to 2050.”

Is that pace of building and innovation even possible in hydrogen, about which many people are extremely skeptical? “If it’s required, it’s doable,” says Friedmann, “and the hydrogen piece is required. There’s no pathway that gets us to where we need to go without gobs of zero-carbon hydrogen.”

As for CCUS and bioenergy, in the FIC, “CO2 capture would increase by 50% to around 7.5 GtCO2 per year in 2050,” it says, “while almost 90 new bioenergy plants equipped with CO2 capture and storage would be needed each year, nearly three times as much as the capacity projected in the Sustainable Development Scenario.”

ccus under faster innovation IEA

(For the latest on the near-term prospects for carbon capture, see the work of Jennifer Wilcox and her team at the Worcester Polytechnic Institute on CCS as applied to natural gas, the US industrial sector, and direct air capture.)

Similarly, the rapid development of battery technology will enable much faster electrification of heavy transportation (and thus the building of new fleets):

vehicle electrification under faster innovation IEA

Building at this pace (two steel plants a month, 90 bioenergy plants a year, and similar numbers for other clean technologies) is virtually unknown outside of wartime, and it has certainly never been done globally. “There is little or no precedent for the required pace of innovation in the Faster Innovation Case,” IEA says, “and it does not leave any room for any delays or unexpected operational problems during demonstration or at any other stage.”

If there are delays — if, for instance, in the US, a Republican Senate refuses to pass any legislation that might politically benefit new Democratic president Joe Biden — the eventual task will grow more difficult and expensive.

“A delay in demonstration projects and a slowdown in deployment of early adoption technologies following the Covid-19 crisis would require greater government efforts down the line,” IEA says. “For example, capital costs of key technologies like hydrogen electrolysers could increase by up to 10% by 2030, making it harder to scale up production.”

To accelerate out of the virus slump, every government needs to get going on clean energy innovation immediately, and at speed.

There’s more than enough consensus on innovation to move forward

Three quick final things to say about the report.

First, I was happy to see IEA emphasize a point I’ve been making for years, which is that smaller-scale, distributed energy technologies, including digital and information technologies, innovate faster than large, capital-intensive techs like carbon capture facilities or steel plants.

rooftop solar
Small and modular.

“When you get to these really large-scale infrastructure projects like CCS,” says Sonia Aggarwal, an energy analyst with the research firm Energy Innovation, “each stair step of innovation is a billion dollars, at least, and you have to build a number of them before you realize what is going well and what isn’t.”

In contrast, IEA points out, smaller and more modular technologies lend themselves to mass production, standardization, and continuous learning, which have benefited solar PV and lithium-ion batteries. Because smaller techs are frequently networked (especially in electricity), they produce large knowledge spillover effects into other technologies.

And digital technologies are not only quick to iterate, they can accelerate the pace at which other, non-digital technologies reach market. Internet and communication technology (ICT) can help substitute computing power, which is getting cheaper, for labor and material, which are not — substitute “intelligence for stuff,” as I like to put it.

Here’s the thing, though. It seems to me that if a) you desperately need rapid innovation in clean energy technologies, b) some clean energy technologies innovate faster than others, and c) you are short on time, it makes sense to lean into the tech that moves faster. Smaller, modular, distributed tech might not get you all the way there — there may never be a small, modular steel plant — but, especially in electricity (the core of the future energy system), it can do a great deal.

A more rapid rate of innovation is a reason to prefer smaller, networked technologies, and to think about how a future energy system can be built around them. Though IEA does not draw that conclusion explicitly, I think it ought to.

Second, IEA is still drawing on the climate models that inform the IPCC and other international climate organizations, and there are some climate analysts who think those models underestimate the growth of clean energy and thus overestimate the future rise in emissions.

“We might be reaching somewhat pessimistic conclusions due in part to limitations of modeling,” says Aggarwal. “We may be making the problem harder on ourselves than it needs to be.”

This dispute matters, because if zero-carbon energy can get closer to true zero emissions, then negative emissions will have less work to do — and R&D should shift from CCUS to electrification and hydrogen accordingly. (Again, it’s not an either-or choice, but there is a question of priorities.)

It makes sense to get CCUS ready for the market regardless, if nothing else as a hedge against downside risk, but if the decarbonization task is going to be less difficult — still incredibly difficult, but less difficult — than we currently project, we can take a somewhat more hopeful and less defensive view.

Put all that aside, though. My third and most important point is simply that, no matter what you might think about the ideal path or energy mix for getting to net zero, there is no credible climate analyst who disagrees with the need for rapid innovation.

Innovation in early-stage technologies is not, as it has sometimes been seen by climate advocates, a substitute for or a distraction from the deployment of existing, mature clean energy technologies. It is obvious that both innovation and deployment are urgently needed. Indeed, as IEA points out, large-scale deployment is rightly seen as a part of innovation policy. Deployment is one of the key forces driving innovation in later stage technologies, through economies of scale and learning by doing.

Whatever their disagreements about the quality of climate models, the merits of various technologies, or the ultimate prospects for success, climate policy analysts agree on the short-term need for radically increased R&D spending, a system for moving technologies quickly over the “valley of death” between lab and market, and a rapid buildout of enabling infrastructure like high-voltage transmission lines, electric vehicle chargers, and CO2 pipelines.

Most countries aren’t doing anything like this. “Low-carbon energy R&D spending in IEA member countries has been broadly stable since 2012, after doubling between 2000 and 2012,” says IEA. “It remains below the levels in the 1980s, however.” That is wild.

us energy r&d spending
In 2016 dollars, the US Department of Energy spent more on renewable energy R&D in 1978 than in 2018.

The task may appear particularly challenging from the perspective of dysfunctional US politics; there are at least a few successes to cite. Anna Goldstein, an energy researcher at the University of Massachusetts Amherst, points to ARPA-E, the advanced energy research agency created under Obama. “There was an idea for innovation policy, it got put into practice, and now we’re seeing results from it,” she says, “and people on both sides of the aisle are saying, ‘this is great, let’s scale it up even further.’ That’s promising.”

For good examples of long-term thinking and planning, she cites the US Mid-Century Strategy for Deep Decarbonization and the Quadrennial Energy Review. “We can pretend we all just blacked out for four years,” she says, “and then build on the successes of the Obama administration.”

We know what needs to be done and, more or less, how to do it. The recipe for responding to climate change is simple: Learn and build, learn and build, learn and build, not as part of any one-time, grandiose mega-plan, not toward any particular finish line in 2030, 2050, or even 3000, but as a way of organizing our collective life on the planet, the central and intense focus of all our wealth and ingenuity.

“There’s nothing in the physics, chemistry, or finance that’s prohibitive,” says Friedmann, “the question is whether we can we learn to tie our shoes.”

Support Vox’s explanatory journalism

Every day at Vox, we aim to answer your most important questions and provide you, and our audience around the world, with information that has the power to save lives. Our mission has never been more vital than it is in this moment: to empower you through understanding. Vox’s work is reaching more people than ever, but our distinctive brand of explanatory journalism takes resources — particularly during a pandemic and an economic downturn. Your financial contribution will not constitute a donation, but it will enable our staff to continue to offer free articles, videos, and podcasts at the quality and volume that this moment requires. Please consider making a contribution to Vox today.

Posts from the same category:

    None Found