Guest Post: Gregory Nemet shares 6 key lessons to inform negative emissions technology innovation

Gregory Nemet, an Associate Professor at the University of Wisconsin–Madison in the La Follette School of Public Affairs and the Nelson Institute's Center for Sustainability and the Global Environment, writes in this post about how the history of other technological innovations can inform our expectations and policy around the development and deployment of carbon removal solutions. 


Meeting the ambitious climate change targets agreed upon in Paris last December will require deep transformation of the global economy—especially in energy systems, transportation systems, and industry—over the next several decades.  It is becoming increasingly clear that such a transition will almost certainly require substantial deployment of negative emissions technologies (NETs) during the course of the 21st century. 

It is becoming increasingly clear that such a transition will almost certainly require substantial deployment of negative emissions technologies (NETs) during the course of the 21st century.

One way to look at this challenge is through the lens of integrated assessment models (IAMs), which are optimization models that minimize the costs of reaching climate targets over the long term.  Even though they have so far included only a subset of potential NETs, these models deploy 5 to 20 gigatonnes (GT = 1 billion tonnes) of CO2 removal per year (global CO2 emissions are around 40GT per year today) in scenarios that correspond to the Paris targets (e.g. limiting warming to +2C degrees).  Deployment of NETs will surely increase as these models start to develop ways to achieve +1.5C degree targets, as the IPCC has been asked to report on.

Integrated assessment modeling from the Global Carbon Project shows negative emissions prevalent across climate scenarios.

Integrated assessment modeling from the Global Carbon Project shows negative emissions prevalent across climate scenarios.

A less black box way to understand the challenge is through carbon budgeting.  Meeting those targets allows the world to emit about 1000 more gigatons of CO2—at current rates we’d reach that limit around 2040 and we’d have to be at zero from then on.  The budget for +1.5C degrees, which also was included in a more aspirational way in the Paris Agreement, would mean getting to zero in the 2020s if emissions were to stay constant until then.  More realistic scenarios include a peak reasonably soon and then smooth decarbonization thereafter.  But the math of +2C degrees, means that peak has to occur very soon and the decarbonization must be rapid, not gradual. 

If we want a more gradual transition, we need to start thinking about a warmer world than +2C or think seriously about negative emissions.  Many possible ways have been proposed to remove CO2 from the atmosphereI found at least six in which peer reviewed journal articles have included estimates of potentials of at least 1 gigawatt of CO2 removal per year.  Some have potentials of 10 GT/year or more.

BECCS: bioenergy with carbon capture and storage, DAC: direct air capture, EW: enhanced weatherization, AR: afforestation and reforestation

BECCS: bioenergy with carbon capture and storage, DAC: direct air capture, EW: enhanced weatherization, AR: afforestation and reforestation

It would be a mistake to interpret this comparison as saying that our capacity for removal exceeds our need.  These are simply estimates.  There may be negative interactions among them so that they do not sum.  Each has potentially serious questions including: competition with food, permanence of storage, energy consumption, cost, public acceptance, and verifiability.  All of these issues merit serious consideration and may limit realistic potentials.  What is a valid insight from this comparison is that a diverse set of possibilities exists.  While it is far too soon to concentrate on any of them, it is also too early to write off any of these methods based on their challenges. 

While it is far too soon to concentrate on any of them, it is also too early to write off any of these methods based on their challenges.

To turn these possibilities into options—that is technologies that we can deploy if we need them—we need a set of policies to accelerate innovation in them so that they become scalable real world technologies.  I’d suggest that designing such policies should start with what we know about historical case studies of analogous innovations and government efforts to encourage them.  Here are a few to begin:

1. Historical case studies show that successful innovations are those that combine technological opportunity with a market opportunity.  Market experience is crucial; it informs new research and incremental improvements via learning by doing and economies of scale.

2. Research and Development (R&D) is needed, but to make these technologies real, look to early deployment, not scientific breakthroughs.  R&D can enable scale up and address challenges, such as in materials, reactions, and storage.  But NETs are not a challenge like the Manhattan- or Apollo Projects, even if it shares the urgency of ending a war or landing on the moon.  The challenge of developing NETs is more like rural electrification, the interstate highway system, and the green revolution.  These involved variation, gradual scale up, integration with a larger technological system, and serving diverse end-users.

3. Scale up is central to the challenge and is not trivial.  Both making larger units and deploying many units take time and continuous improvements that learn from previous efforts.  There are plenty of examples of failure due to scaling up too big, too fast. Iteration and gradual scale up would replicate successful strategies in analogous technologies. 

The Kemper CCS Project shows the risks of trying to scale too big too fast.

The Kemper CCS Project shows the risks of trying to scale too big too fast.

4. Expect dynamic costs and non-linear deployment.  Learning by doing and economies of scale bring down costs.  Deployment is likely to follow an S-curve; slow at first due to technical problems and risk averse adopters; and rapid once scale reached, dominant designs achieved, and reliability proven.  Like many other technologies, expect adoption to be slower than expected in near term and faster than expected in the medium term.

Successful innovation requires rapid iteration at small scales in both R&D and deployment. Via Greentech Media and Bloomberg New Energy Finance

Successful innovation requires rapid iteration at small scales in both R&D and deployment. Via Greentech Media and Bloomberg New Energy Finance

5. Demand for NETs needs to be robust.  For those who invest in innovation in NETs, where do expected payoffs come from?  What if the credibility of policies is weak?  The long time scales involved suggest a boom and bust cycle of interest in addressing climate change, rather than a smooth monotonic increase in action.  Serving niche markets, creating co-products, and hedging across political jurisdictions are ways to make demand for NETs robust to policy volatility.

6. Public acceptance will be crucial for all NETs.  In simple terms, we know that public perceptions are favorable when there is familiarity, involvement in decision making process, and when scales involved are human rather than industrial.  Perceptions are unfavorable when deployment is rapid and adverse outcomes are experienced nearby.  If publics are skeptical, interim failures can become high profile and create insurmountable setbacks

To turn these possibilities into options—that is technologies that we can deploy if we need them—we need a set of policies to accelerate innovation in them so that they become scalable real world technologies

A technology strategy for NETs in the near term should focus on initial deployment and iteration.  It should target learning, intelligent failures, and improvement.  The quantity of CO2 stored, efficiency, and cost are secondary; they are progress indicators, not program objectives.  Later is the time for de-risking the technology and targeting cost reductions.  Look for places where many small units are deployed in real world conditions, rather than a few large installations…even if some units must be large eventually.

NETs are only viable as a defense against rapid climatic changes if many units are deployed at small scale before they are needed.  Without this experience, rapid scale up from lab scale to address an emergency are likely to generate: large technical failures, public opposition, and lock-in to problematic designs. NETs only have “option value” once they have been deployed at a small but substantial level.  In short, an innovation strategy for NETs that learns from the past would include:

  • Build
  • Fail
  • Record
  • Improve
  • Repeat…many times, with a diverse set of approaches, at incrementally larger scale, and in increasingly realistic conditions.

Gregory Nemet is an Associate Professor at the University of Wisconsin–Madison in the La Follette School of Public Affairs and the Nelson Institute's Center for Sustainability and the Global Environment. He is also chair of the Energy Analysis and Policy certificate program

His research and teaching focus on improving analysis of the global energy system and, more generally, on understanding how to expand access to energy services while reducing environmental impacts. He teaches courses in energy systems analysis, governance of global energy problems, and international environmental policy.