Leveling the Playing Field for Open-System Carbon Removal

Spotlighting the growing risks that we optimize CDR for the short run at the expense of long-run scalability

Note: This is the first in a multi-part series about big looming risks and flaws in the way we are collectively pursuing CDR pathway diversity—revolving around “open-system” pathways, quantification uncertainty, and scale—that Dai Ellis and I are writing together and publishing simultaneously on our respective blogs. 

It’s hard to overstate the mountain-climbing challenge in front of us if we are going to start removing many gigatons of carbon from the atmosphere by mid century. 

Among many other things, this Everest will require that we develop and optimize the many potential pathways to durably remove carbon. Innovators are already pursuing lots of CDR pathways and more will follow in the future. Some will prove unworkable. But we need to make sure that where pathways fail, they fail for principled reasons intrinsic to their potential for scalability, safety, durability, low cost, and so on—and that they fail despite having had an equal opportunity to succeed. 

Even at this very early stage, there are significant and growing threats to that equality of opportunity for different CDR pathways. Even more worryingly, the pathways that face the biggest risk of seeing the playing field tilted against them—what some have started to call “open-system” pathways—also represent some of our best hopes for scalability. 

Today’s post kicks off a multi-part series about this danger. Below we define what we mean by open- and closed-system carbon removal pathways, and highlight how open-system approaches are likely to be center stage at gigaton scale because they enhance existing aspects of the carbon cycle that already operate at planetary scale. We argue that adopting an expansive and explicit portfolio-building mentality in how we cultivate CDR pathways needs to be at the core of our scalability strategy. 

Our next post will explore one of the biggest tilted-playing-field risks—CDR quantification uncertainty—and its silver linings. The tail end of the series will discuss how the CDR community might navigate uncertainty from a market-shapingperspective—ultimately distilling some practical guidance tailored to particular market actors (especially buyers) in the interest of establishing the incentives and level playing field that will enable CDR to scale.

Open-system and closed-system pathways

There’s a lot to like about the way the toddler stage of the durable CDR market has started to play out. The talent influx, advance market commitment, explosion of VC funding, and uptick in government support that we’ve seen in just the past couple of years has been remarkable. And most importantly, key buyers and opinion leaders like Shopify, Carbon Plan, and Stripe have helped build alignment around a high bar for the hallmark features we seek in carbon removal methods: safety, verifiability, unimpeachable additionality, extremely high durability, low cost, and so forth. 

Knowing this high bar will be hard to meet, leaders in the CDR community talk about “maximizing shots on goal” by supporting experimentation with lots of different possible methods for durable carbon removal. But in practice, we’re not yet securely on track to give all of these shots on goal an equal shot at success. And the paradox is that the height and firmness of our high bar—a much-needed rigor that we need to preserve at all costs as the foundation for everything—exacerbates the risk of tilting the playing field. The problem is not the high bar itself, but rather that the CDR market may not react in the right ways to the fact that some carbon removal pathways necessarily have a longer path to meeting certain aspects of our high bar.  

To understand how and why, we need to talk about the distinction between open- and closed-system pathways. It’s a useful but not binary divide, because the spectrum from fully open-system pathways to fully closed-system ones is a continuum; any given pathway can have elements of both.¹ Broadly, closed systems—notably pathways like direct air capture (DAC), bio-oil sequestration, BECCS, and some other forms of biomass carbon removal and storage (BiRCS)—generally involve a larger fraction of pathway steps with a high degree of human control and engineering, and critically, carbon sequestration that involves putting a known quantity of carbon in a controlled environment, often a confined geological space. 

Meanwhile, open-system pathways—including enhanced rock weathering and ocean CDR approaches like ocean alkalinity enhancement and oceanic biomass sinking²—involve a larger fraction of pathway steps outside direct human control and sequestration in open natural environments. In open-system CDR, humans are intervening, but also piggybacking on natural biogeochemical processes to do some of the work of capture and/or sequestration. And the further toward the open-system end of the spectrum a given pathway lies, the fewer deployment elements are happening within a contained system and the more are occurring in natural settings—on fields, in our oceans, and so forth. 

Inescapably, open-system pathways come with a thicker set of what we might call “sources of messiness”—risks, uncertainties, fears, hurdles, and so forth. These sources of messiness map directly to extra burdens of proof or layers of work that await open-system CDR researchers and practitioners. A few in particular are worth calling out:

In light of all these sources of messiness, the rub is that even if we did have a perfectly level playing field, open-system pathways have a bigger chasm to cross to meet the rigorously high bar we have rightly set for durable CDR. Particularly as regards three core elements of that high bar—ecosystem safety, quantification certainty, and workable pricing (the challenge on this last one being not cost structure but rather the way that uncertainty-driven discounting increases the effective price per ton)³—many open-system approaches will have an entirely viable path to meeting the bar, but also an unavoidably longer path.  

Open-system pathways and scale

Why bother with all this messiness? There’s plenty of reason to be fearful of unintended consequences with large-scale interventions in nature, given humanity’s track record of destruction. Might it be OK if what we wind up with is a durable CDR industry in which ‘tidier’ closed-system supply predominates? 

One big reason the CDR community should specifically care about open-system pathways is that they have big advantages as we think about long-run scalability.

Zooming out, carbon removal is a mass transfer problem. Moving mass requires a lot of work, in the thermodynamic sense. We are also talking about moving an extremely dilute, fully oxidized atmospheric gas. To do CDR seriously at gigaton scale is to inherently be committing crimes against the laws of thermodynamics. 

The more that we—ie, humans and our machines—have to directly move or manipulate each of the molecules we have to work with, the more extreme our crimes are and the more thermodynamic work we have to carry out and fuel. Adding order to the disordered geophysical system has implications for energy needs, for logistics complexity, and for a mix of other resources and arrangements that need to be lined up to execute the mass moving. Critically, anything that is a required input or enabler becomes a potential constraint to scale and a potential cost contributor. 

Direct air capture has the worst of this, requiring us to "touch" every single molecule of carbon we remove. But by bringing every step under our control, we can directly observe and measure. In essence, closed-system pathways are buying high quantification certainty at the cost of doing additional thermodynamic work on the system. 

Open-system pathways minimize parts of this uphill fighting by enhancing existing natural biogeochemical processes that already remove carbon durably at huge scale. With open-system CDR, we bear a smaller share of the mass-moving burden and nature bears more. This comes at the expense of being able to control and measure each step in the pathway, creating additional quantification uncertainty. But augmenting existing processes in the natural carbon cycle potentially lets us scale far faster and with better energetics and cost structures.

There are several open-system pathways where this scale advantage feels intuitive: 

• Sinking terrestrial biomass in anoxic basins is a matter of transporting pre-captured carbon in the form of waste biomass to large, durable storage sites created by natural pickling processes in these basins. The complex problems of capture and durable storage get simplified to just executing the sourcing, transportation and logistics steps.

• Enhanced rock weathering involves grinding up basalt rock and transporting it to agricultural fields where the spreading equipment often already exists—natural weathering chemistry and ocean storage take care of the rest. 

• At the far end of the spectrum, ocean iron fertilization would involve transporting and deploying relatively modest quantities of micronutrients in the right places, letting the ocean’s massive biological pump take care of the rest.

These kinds of pathways maximize the piggybacking we’re doing and minimize the set of new, hard obstacles a CDR supplier needs to overcome each time it confronts the next big step up in deployment scale. Things like building a new reactor site, raising billions in project finance for high-capex deployments, or negotiating large contracts for 24/7 clean energy are all complex, high-risk tasks that constrain scaling speed. 

Another useful scalability lens is to think more broadly about the piggybacking concept: not just how we piggyback on natural biogeochemical processes, but how we harness existing industries and supply chains. Heavy-industry players—in mining, construction materials, agriculture, shipping, bulk transport and more—will have particularly low-hanging-fruit opportunities to slot into supplying or deploying open-system carbon removal. Waste biomass, rocks, and other alkaline materials at the core of many open-system pathway supply chains are already produced at enormous scale by existing industries. These are only directional observations on a spectrum; but the more a CDR approach can avoid inventing or overhauling supply chains, the faster a path it has to gigaton scale.

The kinds of existing biogeochemical processes that open-system pathways enhance have such a powerful impact on atmospheric carbon and radiative forcing that paleoclimatology suggests they have produced dramatic swings in the Earth’s climate. About 49 million years ago, the sudden gigaton-scale growth and sinking of the Azolla fern brought about the transformation of Earth from a hot, greenhouse planet to our current much colder climate. As another example, natural iron-containing mineral dust is always getting blown into the ocean—but research suggests that periods of greater transport have seen boosted primary production high enough to explain Earth’s transition from interglacial to glacial periods. If such pathways can create change of that climatic scale even without human involvement, they hold intriguing potential for faster gigaton-scale removal with the right augmentation.

Of course, because we’re intervening in natural processes that we don’t directly control, there is a lot of science that needs to be done to constrain uncertainties and mitigate risk to ecosystems. This is the messiness we described above. We have to take these issues extremely seriously, and there’s no way to compare their inherent risks apples-to-apples with the different flavor of scaling risks that will weigh heavier on closed systems. But a primary reason to put up with and work through all this messiness to render some subset of open-system pathways viable is that—where they prove safe and effective—they could reach much greater removal volume much faster. 

The advantages of greater piggybacking on thermodynamic work done by nature may not matter much if the long-run goal were just just megatons, but since the goal is gigatons they are likely to be crucial. As one early example, we already see signs of this scalability in medium-term deployment curve projections among ERW suppliers.⁴ If we let open-system carbon removal possibilities wither on the vine we are placing a huge and risky bet on the scalability of closed systems. 

A “portfolio” mentality for scaling CDR 

Thermodynamic piggybacking is one reason to care about open systems and carbon removal pathway diversity. But it’s far from the only one. More generally, we want to argue that there is great long-term benefit in cultivating the broadest possible portfolio of viable pathways because pathway diversity is valuable in itself.  

When people talk about “maximizing shots on goal” they generally mean that we’re experimenting with many CDR pathways because not all will work out. It's far too early to definitively conclude which subset of pathways will end up having the most attractive profiles from the perspectives of cost, scalability, energy and land usage, environmental impacts, and so on. Pathways that may seem like emerging winners now may turn out to be flat out be infeasible at the gigaton scale and play a negligible or niche role. 

But we aren’t pursuing all of these in the hopes that just two or three of them will be massively scalable. And in this sense the “maximize shots on goal” mantra—implying many shots and few goals—is misleading. We’re pursuing many pathways because we will need many. Why?

Partly, this also is about scalability. We have a better chance of reaching astronomical scale if we have many companies pursuing many pathways with many technologies. But parsing that a bit further, what we particularly need is to diversify scaling risks and scaling constraints by growing many pathways in parallel. In other words, we should prefer a world in which different pathways are encountering different problems as they scale. We want to avoid a world where the dominant constraint becomes access to permitted wells or access to 24/7 cheap clean electrons in a few key geographies or a particular workforce bottleneck; or where the overriding risk is growing public backlash about a particular environmental risk. These are brittle futures: concentrated risks and constraints that threaten to slow progress dramatically. 

We are better off if we have some pathways that are limited by biomass feedstock availability, others by nutrient availability, others by total energy demands, and so on. Constraints and risks to scaling are just as certain as death and taxes, but we should want them to be diffuse and diversified rather than narrow and concentrated. 

Pathway diversity also helps in socioeconomic ways. We want a massive CDR industry to be highly inclusive and equitable, to create a diversity of jobs, and to harness as much human talent and effort across the globe as possible. Pathway diversity minimizes the chances of a situation where scaling gets slowed by our ability to train people in one or two key trades. But framed more aspirationally, pathway diversity means that we can equally well create a new CDR job or income stream for an electrician or a farmer or a marine technician or an oil & gas worker. That we can establish large-scale CDR supply in places as diverse as Kenya, Oman, India, and Namibia even though each has a unique set of natural and human resources that is well-suited to some CDR pathways and not others. And, thereby, that we can create a future in which trillion-dollar CDR is supporting broad-based prosperity and not just exacerbating inequality. 

Broad-based prosperity is something we want in its own right, but it’s also one of the most central ingredients in the scalability recipe. The only way we successfully build global popular and political support for gigaton-scale deployment and trillion-dollar CDR spending is if that spending produces broad-based prosperity. Carbon removal becomes an important new international economic engine in an era of slowing population growth. Deployment creates geographically-distributed jobs and wealth and tax revenue that in turn fuel political support for more deployment. 

The growing specter of a tilted playing field

So we need to build a broad portfolio, helping the best open- and closed-system pathways get over the hump and flourish over time. This means we need to maintain a level playing field across pathways that hold clear potential to meet our high bar and scale quickly. What do we mean by a level playing field?

While you could certainly add things to this definition, in our mind the most important components of a level playing field are:

1. Buyer behavior: the market—meaning private companies in the near term, and government in the medium term—does not discriminate among pathways in arbitrary or unprincipled ways, and buyers show consistent willingness to enter into offtake agreements with any pathway that starts to meet our high bar (noting that in the case of cost this will often mean meeting expectations for future cost potential). 

2. Government policy:

   1. Government incentives are technology- and pathway-agnostic in the sense of comparably incentivizing investment in different pathways with similarly attractive long-term potential.

   2. Government policymaking is not captured by deep-pocketed lobbies that steer policy in favor of their preferred CDR pathways.

3. Prevailing narratives: the narratives that take hold within the CDR market and community—about which pathways have what kind of future potential and are therefore worthy of investment, demand and policy support—are consistently rooted in evidence and not overly short-term-minded. 

In aiming for a level playing field across open- and closed-system pathways, the trick is that we have a big difference between the two categories in their ‘messiness’ and therefore in how much learning and work will be needed to fully meet the high bar that we need to uphold. 

The trillion-dollar question becomes how we manage that tension. One could imagine three different stances the CDR community and policymakers could take:

• An “extra support” stance: in this scenario, we would commit to maintaining a level playing field for open- and closed-system pathways as a baseline, as defined above. But in making philanthropic and even public investments to advance science and accelerate field-building work—and potentially certain other forms of support—we would also seek opportunities to steer extra support to open-system pathways in recognition of the extra hurdles they face. Think of it as an analogy to affirmative action in university admissions. 

• A “scrupulous neutrality” stance: here we would simply commit to maintaining a level playing field in all regards, and otherwise invest similarly in open- and closed-system approaches, letting the chips fall where they may. 

• A “lean away from the messiness” stance: in this world, while we wouldn’t try to actively discriminate against open-system pathways, we would tacitly allow a titled playing field to develop if that is the organic result of decisions made by buyers and governments optimizing for the shorter term. We would effectively say to ourselves that the high bar matters more than anything else, and some pathways have a quicker route to meeting that bar, so it makes sense that our purchases and policies end up disproportionately investing in those pathways. 

The early patterns we’ve seen to date suggest that we are drifting toward the last of these potential stances. At the moment, the world is going hot and heavy after closed-system pathways precisely because they are cleaner and tidier. They don’t suffer from messiness like ecosystem risks and higher levels of quantification uncertainty. They have a shorter route to meeting all the elements of our high bar. And they’re also easier to understand, and to describe to policymakers and the public: take a minute and try to explain DAC to someone, then take a minute and try to explain electrochemical ocean alkalinity enhancement. It’s easier to enlist support for tidy things that are easy to grok.   

Where and how do we see hints of an emerging tilted playing field? For one, in prevailing narratives: closed-system pathways like DAC and some BiCRS approaches have simply attracted disproportionate publicity and mindshare in the first few years of the durable CDR market. These pathways have been floating around as ideas for much longer in the academic and policy spheres and have more prominent thought-leader proponents. Because of the faster path to meeting our high bar, the closed-system arena also sees more headline-grabbing commercial ventures and milestones (e.g., CarbonCapture’s Project Bison and Charm Industrial’s industry-leading early deliveries and new Climeworks plants and the 1PointFive/Airbus deal). 

As one would expect, the state of the public conversation is reflected in the state of public policy. Taking just the US as an example, the primary CDR-accelerating tax incentive (the 45Q credit) applies only to a subset of closed-system pathways, and the lion’s share of direct government investment announced to date (notably the $3.5 billion for DAC Hubs) skews in the same direction.⁵ And the quicker commercial traction with closed-system pathways fuels the cycle, as deep-pocketed industrial interests like Oxy who see big near-term opportunities lobby hard for more policy provisions and government funding channels slanted even further in favor of closed systems. Meanwhile, policy advocacy organizations naturally focus on early wins they can point to and the pathways that are easiest to talk about, while environmental NGOs who support mainstream ideas like afforestation or mangrove conservation balk at newer ideas like large-scale OAE. 

These risks are coalescing even before we fully confront what may become the biggest source of tilted-playing-field risk in the coming years: how buyers and others navigate quantification uncertainty. Carbon removal via open-system pathways is generally subject to greater quantification uncertainty because capture and sequestration are the sum of a complex system of interrelated carbon fluxes—ones that generally occur beyond our direct control and observation, over time, and across large terrestrial or ocean areas. By comparison, the amount of carbon removed from the atmosphere via closed-system pathways can be relatively easily measured because you directly and observably transfer the carbon you have captured into what amounts to a tank. There may be some moderate sources of uncertainty, such as counterfactuals related to the sources of biomass used in BiCRS pathways, but generally these are manageable.

This looming honker of a risk is the one we want to unpack most deeply in this series. We will dive into the nuances of quantification uncertainty and discounting (its sibling) more fully in the next post, building on some great recent work by Frontier and CarbonPlan. But for now, suffice to say that quantification uncertainty creates both a bigger MRV challenge and a bigger pricing challenge for suppliers in the near term—and thereby could lead buyers to act very tentatively in purchasing tons from open-system CDR suppliers. 

What this all adds up to is a risky confluence of system dynamics, where the combination of the high bar we’ve rightly set and the relative tidiness of closed-system pathways could tilt the playing field more and more heavily in favor of closed systems over time. What might this bad-movie scenario look like? It would be a world where buyers are acting skittish about uncertainty, steering most of their demand toward closed-system suppliers. Where there’s a growing public narrative and a chorus of opinion leaders and journalists equating CDR with DAC and closed-system BiCRS. Where already DAC-leaning government support doubles down in favor of closed-system pathways. And where money, talent and buzz accordingly follow the direction the current seems to be flowing.

Outlining the way forward

Summarizing two particularly important and interconnected ideas arising from the lines of thinking above:

1. We need to approach carbon removal as if we are collectively constructing a portfolio of pathways. Just like any investment portfolio or other type of constructed portfolio—as distinct from a loose or haphazard assemblage—we need to develop, shape, re-balance, and prune our portfolio as we learn more about the advantages and risks of different pathways. And the diversification within our portfolio has intrinsic value. 

2. We can’t take for granted a technology-agnostic level playing field for CDR that will optimize our long-term portfolio. We have to resist the drift toward prematurely tilting the playing field by preferentially supporting closed-system pathways just because they’re early leaders out of the gate, or getting the most buzz, or feeling the safest and tidiest to get behind.⁶ Avoiding acceleration of that trend will take an intentional, concerted push over many years. 

What are some of the biggest things we need to do in the coming years if we want to set both open-system and closed-system pathways up to thrive and pursue a truly diverse portfolio?

First, we have to greatly accelerate field building and the pace of scientific progress on open-system approaches. We need to step on the gas in tackling quantification uncertainty, understanding ecosystem risks, attracting new talent, developing efficient ways for researchers and companies to secure permits, and much more. At present, this kind of work is both heavily undercapitalized and poorly coordinated, with many siloed ecosystem actors tackling their own small piece of the overall problem.

Second, we need to continuously shape the market and thereby our portfolio. If we believe in the myth that the market will just do its magic, we are much less likely to see open-system pathways flourishing over the long run. As we will start to discuss in the coming posts, we particularly need to get early market signals right: emerging pathways are most vulnerable to tilted playing fields in the initial and adolescent stages of their development when they need support and demand to work through their “sources of messiness.” We’ll need to get buyers and verifiers aligned on principled approaches to uncertainty and discounting; influence buying patterns across a range of types of private and public purchasers; help the CDR community build an updated and more widely shared mental model around these issues; and likely consider targeted forms of innovative-finance interventions to tackle specific bottlenecks and opportunities with open-system pathways. 

And last, the CDR community will need to step up efforts to influence public policy and public narratives to help create a more level playing field. Part of this is ensuring that we have enough advocacy horsepower to influence national policies, in the form of industry associations and policy organizations like Carbon Gap and the newly announced Carbon Removal Alliance. Counterbalancing the political weight of fossil fuel interests, who will tend to lobby for support to closed-system pathways, will not be easy—but one of our best hopes over time will be identifying heavy-industry companies who can play a big role in open-system pathways and getting them more quickly to the policy table. And eventually, we will need much broader-based campaigns to engage the public and influence the currently fear-based conversation around open-system climate interventions. We must convince people that responsible science can help us identify pathways that need to get nixed and scale the ones that can be deployed safely. 

* * * * * * 

Our drift toward a tilted playing field isn’t a conscious CDR community decision, but as a trendline, it’s also not one that’s yet raising a big hue and cry. It deserves more scrutiny. If we implicitly treat closed-system pathways as the bulk of the “real” durable CDR market in the near and medium term, we risk creating a self-fulfilling prophecy that leaves us with a much thinner and less scalable portfolio as we hit the real inflection points in future deployment.

1. Consider, for example, Brilliant Planet's microalgae-based CDR pathway. It has significant closed-system elements in that microalgae are cultivated in tanks near the coast, and the resulting biomass buried underground in the desert. All of this can be directly observed and measured. But the atmospheric carbon removal happens out in the open ocean, when the CO2-depleted seawater from the tanks (having been depleted by the microalgal photosynthesis absorbing carbon from the seawater previously pumped into the tanks) is returned to the ocean and re-equilibrates with the atmosphere. Thus the actual carbon removal is not directly observed and measured.

2. Note that because we are talking exclusively about high-durability carbon removal in this series, we are excluding from consideration lower-durability forms of open-system CDR such as afforestation or soil carbon sequestration.

3. This distinction is really important. While some open-system pathways are likely to have very attractive cost structures in the long run—for the thermodynamic reasons outlined further below—the countervailing factor is discounting. As we will discuss in more more depth in the next post, higher quantification uncertainty with open-system pathways will translate into higher discount factors, which in turn increases the effective price per ton even though it has no impact on intrinsic deployment cost. And because uncertainties will be greatest in the short run, this price impact is likewise bigger in the near term than it will be in the long term.

4. Some ERW suppliers are expecting to hit megaton scale as soon as 2025. In a world with tilted-playing-field dynamics that mostly work against them, this is one clear advantage that some ERW suppliers will have: they will be among the few able to deliver in large volume in the early years of a supply-constrained market.

5. The risks of a tilted playing field exist even within the closed-system category—e.g., bio-oil sequestration not yet being 45Q-eligible—and need to get addressed urgently. Yet these are likely to prove smaller in the coming years than the risk to open systems.

6. To be clear, we don’t actually want a playing field that is naively or blindly pathway-agnostic. Years from now, we want the playing field to evolve to discriminate in evidence-based ways as the thickening evidence base makes clear that we should select for or against certain pathways—which is why the definition above makes reference to evidence and avoiding arbitrariness and meeting the high bar. How to shape the evolution of the durable CDR playing field in the right ways over time is a very nuanced issue and we will have more to say about this later in the series.