The Trillion Dollar Question of Carbon Removal

Note: We are, at best, observers of the CDR space. The below represents observations and opinions, all of which are surely incomplete and very much up for debate. In other words, please don’t @ me (except for constructive dialogue of course!). That said, if you are working on an approach (in R&D, a startup, a non-profit, government or philanthropy) and you think it fits in here, we’d love to hear from you so please reach out! 

Scaling Carbon Dioxide Removal (CDR), to have any hope of achieving net-zero by 2050, is one of the most important, and most difficult, challenges of our time. Obviously rapid global decarbonization of energy, transport and industry is beyond critical, but the IPCC no longer includes any scenarios that keep warming below 2C without significant CO2 removals1. CDR is not a “nice to have” but a “must have” if we have any hope of keeping climate change below catastrophic levels, or of restoring a “safe climate” by the end of the century. 

Current projections assume that achieving net-zero emissions by 2050 will require us to scale CDR so that something like 10 billion tons (IPCC range is 5-16) of CO2 (or equivalent) is being removed from the atmosphere every year (10 GT/year is roughly the annual emissions of China, which is almost a third of the global total). After 2050, bringing atmospheric CO2 concentration back down to a “safe level” (say 350 ppm) will probably require another thousand or two GT to be removed (an additional 20-40 GT/yr if we want to achieve that target by 2100)2. 

But the CDR sector faces some major questions that need new thinking and solutions. Existing techno-economic assessments usually conclude that we should be able to get down to around $100/ton by 2050 (a recent LLNL report pegs it closer to $1303). This is almost certainly less (probably much less) than the social cost of carbon4 (a useful but imperfect measure of the expected marginal cost of each ton of CO2 emitted), so that’s good. But it still implies that we’ll need to create a new industry, largely from scratch, doing >$1T/yr in revenue (about 1% of global GDP) in just 25 years. That might not sound like a crazy tab to address (a really important part of) the climate crisis, but consider that its about ⅓ of global GDP growth and that there are some other big ticket priorities in the coming decades (like decarbonizing the entire global energy and transport systems while simultaneously lifting another billion or two people out of poverty). 

[This] still implies that we’ll need to create a new industry, largely from scratch, doing >$1T/yr in revenue (about 1% of global GDP) in just 25 years.

These assessments have to make fairly optimistic assumptions5 about exponential growth to have any chance of hitting 10 GT/yr by 2050, and there is little margin for error or delay. The CDR sector has made an incredible amount of progress in recent years, thanks in no small part to billions in demand-pull (or market-shaping) efforts such as Advanced Market Commitments (AMCs)6. AMCs (and similar tools) have driven a massive amount of innovation in CDR in a very short time, helping to launch hundreds of startups (the AlliedOffsets database is tracking 835 CDR companies at the time of writing) that are leveraging billions in early-stage investment ($1.2B in 2023 alone7). Not-for-profits like the Carbon to Sea Initiative, Cascade Climate and Ocean Visions (just to name a few) are doing incredibly valuable work to address non-commercial challenges (for example in policy, shared R&D bottlenecks and standards) and to accelerate various parts of the ecosystem. Spark Climate Solutions is helping lay the research foundation for atmospheric methane removal, another key part of the drawdown challenge that has its own unique hurdles to overcome. Methane removal (and, while we’re at it, nitrous oxide removal8) is nascent enough that it’s difficult to include in a framework like this, but there’s no doubt that it could be an important (possibly essential under certain conditions) part of the puzzle and should be urgently explored. 

Despite this incredible progress, the CDR sector needs much more collective action by investors, policymakers, philanthropists, and others to continue to scale. This is, in large part, due to a lack of sufficient progress in the regulatory carbon markets that are essential for driving real demand scale. There are certainly signs of progress on the international stage (from IPCC talks9 and EU efforts for example10), but regulations are not going far or fast enough to achieve the price-points necessary for most of the durable long-term removal technologies that the world needs. Many (maybe even most?) CDR start-ups are struggling to find that next scale of financing (often to build first-of-a-kind pilots and demonstration scale facilities), because investors and banks see little sign that the necessary regulations and demand will materialize, making the perceived risk, and thus the cost of capital, very high. A lack of pilot- and demonstration-scale facilities has even made it hard for the big AMC funds, like Stripe Climate’s Frontier Fund, to roll out purchases as quickly as they’d like. Without a clear plan for the medium-term, there’s a significant risk of a large fraction of these startups disappearing in the next few years, which could have a chilling effect on capital in the space for many years to come and create significant delays in scaling that we absolutely cannot afford. 

Despite this incredible progress, the CDR sector needs much more collective action by investors, policymakers, philanthropists, and others to continue to scale.

So what can we do? 

We’re not here to tell you that all is lost or that we should reduce the scale of our CDR ambitions. Instead, we argue that we need to increase our ambition, doubling down on new solutions and platforms, setting cost targets significantly lower than 100$/ton by 2050 and rebalancing capital portfolios (public, philanthropic and private) to focus more on scalable solutions and platforms that have the best chance at unlocking <$50, or even <$10, per ton. 

What are some ways we can imagine getting to these ambitious targets?

1. The “Super-Abundance” Playbook: Make electricity clean, available and damn-near free.

We could hope that ongoing work leads to some breakthrough in clean energy (dirt-cheap solar? a nuclear fission renaissance? a fusion breakthrough? space solar? geothermal everywhere?) that will make electricity very very cheap. There are already some DAC and CCU companies that are counting on super-abundance as a key part of their business model11. Is it a smart play? Wouldn’t count it out completely, but we definitely don’t think you should bet the farm (and coasts and cities and glaciers) on it. There’s certainly a lot of innovation in the energy sector, but this is big physical infrastructure we’re talking about, and that takes land and time and lawyers and money and a workforce to build out. 

Is it a smart play? Wouldn’t count it out completely, but we definitely don’t think you should bet the farm (and coasts and cities and glaciers) on it.

So what? There’s already a huge amount of investment in low-cost clean energy, so not sure we have a ton of actionable insight here. But there are other (non-technological) bottlenecks that slow down (and add significant costs) to green energy projects, but don’t receive nearly as much attention or funding as the big-ticket high-tech breakthroughs. A big one (particularly in the US) is increasing local resistance to development driven by political opportunism and corporate (i.e. fossil fuel) interests (plus just old-fashioned NIMBYism), which can manifest as small but vocal opposition at the local level that can stop projects in their tracks. Combatting this requires networks and grassroots organizing at a large scale. A fairly new effort called Greenlight America (from some of the same amazing people12 that created the Indivisible Project) is trying to do just that. 

2. Co-benefit solutions 

Many active CDR solutions, particularly nature-based ones, have potentially significant co-benefits such as improved agricultural yields (increasing farmer incomes and resilience), improved coastal resilience or ecosystem restoration. Currently these co-benefits are not “priced in” to carbon credits so they do little to improve the cost profile or attractiveness of a CDR solution (beyond potential marketing strategies). If we valued these co-benefits sufficiently, they could decrease the effective cost of CDR. For interesting examples take a look at what the Penn State PlantVillage project is doing with biochar in Sub-Saharan Africa, what Reefgen is doing to build underwater automation for seaweed cultivation and coral restoration or the approach Vesta is taking to do Enhanced Rock Weathering (ERW) on beaches while improving coastal resilience. While it’s still early days, Mati Carbon deployments of ERW with smallholder rice farmers in India (some living off as few as 1 or 2 acres of land) have demonstrated co-benefits including improved yields, soil health, water holding capacity and pest resistance. Combined with direct revenue-sharing from credit sales, the benefits of more production with lower input costs (e.g. for pesticides) and improved drought tolerance mean farmers can see average annual incomes increase as much as 20% (and hopefully even more). If these sorts of results can scale, it has huge implications for local economies and increased resilience for the very communities that are the most vulnerable to climate change. 

These approaches represent a climate two-for and are severely underfunded (from all sources) relative to their potential.

So what? Hard questions abound, such as how to measure, verify, pay for, and incentivize these co-benefits. And even if we can, it’s far from clear that this would be enough to crack the code. Even so, these approaches represent a climate two-for and are severely underfunded (from all sources) relative to their potential. Philanthropies in particular should be thinking (and some are, but not nearly enough) about how they can solve some of these issues and catalyze larger public programs and private investment. 

3. The Magic O&G Industry Swap (Oil out->Carbon in)

At 100$/ton, we need to build a 1T $/yr industry by 2050. This is like building the equivalent of the oil and gas industry in just 25 years (except again, this industry won’t actually have a “product” to sell in any conventional sense). Coincidentally, it also so happens that we need to more-or-less eliminate fossil fuels in the next 25 years. Can we get a two-for by pushing/incentivizing O&G to transition their vast resources (capital, knowledge and infrastructure) to removing carbon from the air rather than extracting it from the ground13? The sector is certainly big enough (O&G revenues averaged around $2T/yr globally in recent years, with an industry market cap around $100T), but getting the right combination of carrots and sticks is certainly a tall order (short of nationalizing all the big oil companies, but we’ll go ahead and put that in the category of “unlikely”). There are some signs of action on this front, such as Occidental Energy buying DAC firm Climate Engineering for $1.1B last year14, but the economics are not even close to driving the kind of exponential transition the world needs. 

Delayed action on ambitious climate policy means we’re in official crisis mode. It’s all-hands-on-deck and we need every tool in the toolbox and every dollar in the bank.

So what? Obviously governments need to be thinking about how to design creative transition solutions like these over the next few decades. But will they? Philanthropy should think about how they can catalyze the necessary political change by addressing some of the tough challenges now that make this transition “feel impossible”. Unfortunately many (maybe most?) climate philanthropies find the idea of including O&G as “part of the solution” to be anathema to solving the climate crisis. It’s a stance that we certainly have a lot of sympathy for, and probably would have actively supported 20 years ago, but delayed action on ambitious climate policy means we’re in official crisis mode. It’s all-hands-on-deck and we need every tool in the toolbox and every dollar in the bank. Adversity makes strange bed-fellows. There are no atheists in fox holes. Everyone has a plan until they get punched in the face. (Yes, we know some of those don’t really make sense here. But you get the point.)

4. Co-product solutions w/ stackable platforms 

What if we could make the economics more attractive by “stacking” technologies on the same platform(s) to produce multiple products, one of which would be carbon credits? Co-products here could be green electricity (in cases where CDR is “stacked” on a powerplant) or things like building materials or polymers15 (in cases where the captured carbon is being utilized to create products that support long-term durable storage16). There is also the potential to combine Ocean Alkalinity Enhancement (OAE, a promising approach on its own but with significant energy, supply chain and measurement/verification challenges that need to be solved before costs can really come down to where they need to be17) with hydrogen production18, though we’re not aware of anyone trying this at scale and the technology and economics still have a ways to go. 

One of the most interesting approaches in this category is geothermal heat/electricity stacked with carbon capture (and potentially even methane and nitrous oxide capture). There are many advantages to this approach. For instance, most direct air capture (DAC) approaches require significant amounts of heat and getting this heat directly from the earth is vastly more efficient and cheaper than producing it from clean electricity. The geothermal platform also gives you options for storage through something like injection19 and mineralization20. Because of the accessibility of geothermal energy, Iceland has become a major testbed for innovation in this space, with companies like Carbfix opening FOAK facilities in recent years21.

Unfortunately geothermal isn’t available everywhere, but there are exciting efforts underway to try and change this, particularly through the efforts of Project InnerSpace [disclosure: JE recently joined InnerSpace as the VP of Innovation], dozens of next-gen geothermal startups, and renewed attention on the technology from governments around the world22. 

There may be an interesting hack here by combining geothermal-based stacking with the O&G swap.

Still big questions about whether this category can drive down costs enough for the economics to make sense, and whether the approach can scale given the massive technical, capital, and knowledge resources necessary. But there may be an interesting hack here by combining geothermal-based stacking with category 3 (the O&G swap). Finding and exploiting geothermal potential shares a lot in common with finding and exploiting oil and gas (most obviously the drilling and fracturing). Many big oil and gas companies already have geothermal development teams even, but rarely is geothermal anything close to a priority (the economics just don’t compare to oil and gas) so these groups are rarely resourced effectively. Changing the incentives and driving geothermal innovation and scale in O&G is a key part of the mission of Project InnerSpace, which recognized this opportunity many years ago23. 

It’s an ambitious play but it’s another two-for and has the potential to drive real systemic global change.

So what? Can we get to realistic economies for CDR by stacking it with geothermal, simultaneously producing clean heat and electricity, and achieve scale by incentivizing the O&G industry to double down (or really 10-100X down) on geothermal and co-located CDR? It’s an ambitious play but it’s another two-for and has the potential to drive real systemic global change so we think it’s more than worth a shot. Doing it at speed and scale will require ambitious societal shifts at every level: culture, policy, technology, finance, legal. Project InnerSpace is an important start (and one I’m particularly excited about obviously) and is making amazing progress, but this pathway still needs significantly increased support from all financing sectors if it’s going to succeed. 

5. Nature-based solutions with the potential to scale at much lower cost

We’ll divide this category into two: 

5.1. Novel applications of popular chemical methods that avoid big cost categories 

Examples here include things like the startup Aquarry, which proposes to do alkalinity enhancement (AE)24 in abandoned mine quarries (rather than in the ocean or on farms). This has the benefit of significantly reduced Measurement, Reporting and Verification (MRV) costs, basically because each quarry is a closed system. There are also few environmental concerns (probably more likely to be benefits than risks since these quarries are generally highly toxic and acidic) which should significantly reduce both regulatory and supply-chain costs. AE requires sourcing massive amounts of finely crushed rock (basalt in most cases) which can contain trace amounts of toxic metals. Most applications require finding (and testing) highly pure sources, but obviously this is much less of an issue if the target system is highly toxic to begin with. This could reasonably get CDR costs down closer to ~$50/ton (or less), but there are limits to scalability.

If something like 5-10% of global quarries could be exploited, this could make a real dent in the 2050 target.

There are a lot of abandoned mine quarries in the world (maybe about 10,000) with an estimated total capacity >100 GT CO2. Dealing with 10,000 landowners (and the relevant regulators) is a daunting task, but the approach probably needs <100 sites to achieve a billion tons (total) of removals. If something like 5-10% of global quarries could be exploited, this could make a real dent in the 2050 target, and if we could find a few more approaches with these sorts of attractive economics, then the math might just work out. 

So what? This one’s pretty simple. We need to take more (and better) shots on goal (in general, but the private finance sector in particular). We need to make sure we are encouraging out-of-the-box thinking and not getting too stuck on a few pathways. Governments need to think about how they can facilitate this kind of innovation by making it easy to do testing and development on the vast amount of government-owned lands (and they should probably rebalance their investment portfolio to put a whole lot more money into solutions like this and a bit less into $1500/ton DAC plants…). Philanthropy can obviously do more to help to jumpstart these innovative solutions and should invest more broadly in facilitating plays (the analog of the Carbon to Sea initiative for solving big ecosystem and innovation challenges that don’t have clear commercial drives). 

5.2. Known (though still challenging) biological pathways on land or in the ocean 

Examples here include trees (tried and true but with hard-to-measure externalities and longevity concerns) or seaweed (cost and scaling challenges). But perhaps the most promising approach is known as Ocean Iron Fertilization (OIF). Many people prefer the more general term Ocean Nutrient Fertilization (ONF), in part because it sounds less scary to the general public but also because there are some more complex proposals that involve adding other nutrients into the mix in addition to iron, to try and optimize the effect and reduce externalities. It’s basically the same party, just a bigger guest list.

OIF has been studied actively for about 3 decades and has the potential to be very cheap (<$10/ton25), scalable and rapidly deployable. This is because there are large parts of the ocean where the limiting factor for phytoplankton growth is an extreme deficiency of iron. Since the tiny plants only need trace amounts to grow, a little bit of iron goes a long way (1 ton of iron could sequester 1000 tons of CO2, and possibly much more, though estimates vary widely and there are many uncertainties and dependencies on local conditions26). This theoretical potential famously led John Martin, who originally proposed the approach, to make the evocative statement “Give me a half tanker of iron, and I will give you an ice age.”

“Give me a half tanker of iron, and I will give you an ice age.” — John Martin

Like many technologies that seek to tweak natural processes to mitigate climate change, OIF has natural analogues that go some way to validating its potential effectiveness and safety. The Atlantic Ocean for example is a poor place for OIF because it already receives large amounts in the form of iron-rich dust that blows off the Sahara desert (rivers contribute a lot too). Volcanoes also produce iron-rich dust and ash that similarly “fertilizes” large areas of ocean. 

Unfortunately the field has been plagued by controversy, mostly because of the actions of a few bad (commercially focused) actors27, causing funding for research to completely dry up for more than a decade. And there continue to be valid concerns about the method’s effectiveness (which are similar to the concerns for most, and probably all, bio-mediated CDR pathways such as forest management), including the durability of the sequestered carbon (how much ends up back in the atmosphere after 30 years? 100?) and spillover effects like the potential for nutrient redistribution (basically that enhanced phytoplankton blooms in one area could redistribute nutrients in a way that could reduce the net CO2 removal effect of the added iron28). There are also concerns for ecosystem impacts and externalities that require further study, including potential toxic algae blooms and increased eutrophication (a recent National Academies report did not find much evidence for these concerns, but more research is needed29). 

Recently an international collaboration of experts called Exploring Ocean Iron Solutions (ExOIS) has been working to reinvigorate research into this potentially important option. The ExOIS Paths Forward Report is a damn-near comprehensive plan for a responsible but urgent R&D approach to address the remaining scientific questions and technical challenges and study the environmental concerns in order to prove (or disprove) the safety, efficacy and scalability of carbon removal via OIF. The timing here is critical. Mostly because we simply don’t have much time to spare of course, but also because many of the scientists that drove the first wave of OIF research are now approaching retirement. Losing the knowledge and experience of these scientists before a new generation can be trained and tested would represent a huge loss for the field. 

So what? No matter what you think about the risks of OIF, the fact is that it remains, after all these years, the best candidate for a CDR solution that is 

1. cheap enough to be practical under many reasonable economic and political assumptions,
2. has the potential to achieve a scale commensurate with the scale of the challenge,
3. and can be rolled out quickly should the world need these sorts of emergency options. 

For these reasons, we must do the necessary science and engineering to determine if OIF can be done and scaled effectively and safely. And it’s essential that we start now so that these questions can be answered before it’s too late. Since governments are unlikely to move quickly on this, we believe that philanthropy must (and can) urgently step in to fill the gap. 

[Scaling CDR] will require an all-of-the-above approach, a willingness to think and act big, and a “risk-risk” perspective.

All the categories here have merit to varying degrees, and each has aspects that are not receiving sufficient attention and funding, so this can’t be a zero sum calculation. Again: Scaling CDR to have any hope of achieving net-zero by 2050 is one of the hardest and most important challenges of our time. It will require an all-of-the-above approach, a willingness to think and act big, and a “risk-risk” perspective that considers both the risks of action and of inaction and takes the steps now to understand and mitigate those risks by advancing the best available science. 30 

---

References

1 https://www.carbonbrief.org/analysis-what-the-new-ipcc-report-says-about-how-to-limit-warming-to-1-5c-or-2c/

2 https://www.metoffice.gov.uk/research/news/2018/how-much-co2-at-1.5c-and-2c

3 https://roads2removal.org/resources/

4 https://www.nature.com/articles/s41586-022-05224-9 & https://www.nature.com/articles/s41558-018-0282-y

5 https://www.sciencedirect.com/science/article/pii/S2590332223003007

6 For example, see https://frontierclimate.com/writing/launch

7 https://www.cdr.fyi/blog/2023-investment-landscape-in-carbon-removal

8 https://link.springer.com/article/10.1007/s11356-016-6103-9

9 https://www.iisd.org/articles/deep-dive/will-international-carbon-markets-finally-deliver

10 https://www.reuters.com/sustainability/eu-step-up-efforts-more-carbon-markets-worldwide-2024-02-13

11 https://terraformindustries.wordpress.com/2023/01/09/terraform-industries-whitepaper-2-0/

12 https://www.linkedin.com/in/matthew-traldi-82236411a

13 https://www.economist.com/the-world-if/2020/07/04/what-if-carbon-removal-becomes-the-new-big-oil

14 https://www.reuters.com/markets/deals/occidental-petroleum-buy-carbon-engineering-11-bln-2023-08-15

15 Notably, this doesn’t include cases where GHGs are used to create products like fuel, because the carbon ends up back in the atmosphere pretty quickly.

16 https://www.sciencedirect.com/science/article/abs/pii/S175058362100061X

17 https://www.edf.org/sites/default/files/documents/Ocean%20Alkalinity%20Enhancement.pdf

18 https://www.sciencedirect.com/science/article/abs/pii/S0360319923024229

19 Injecting CO2 can be difficult depending on the geology – the main risk is calcites or other solids reacting to form solids which plug up a well or reservoir.

20 https://www.sciencedirect.com/science/article/pii/S0009254120301674

21 Orca plant (https://www.carbfix.com/orca–taking-direct-air-capture-and-storage-to-the) & Coda terminal (https://www.carbfix.com/codaterminal)

22 E.g. https://www.energy.gov/articles/doe-unveils-roadmap-next-generation-geothermal-power

23 https://energy.utexas.edu/research/geothermal-texas

24 https://en.wikipedia.org/wiki/Enhanced_weathering

25 Estimates over the years have ranged from <$1 all the way to $400/ton, so clearly much more research is needed to reduce critical uncertainties and engineer methodological improvements. See for example https://www.sciencedirect.com/science/article/abs/pii/S0928765511000261 

26 https://eartharxiv.org/repository/view/5272

27 https://www.nytimes.com/2012/10/19/science/earth/iron-dumping-experiment-in-pacific-alarms-marine-experts.html

28 https://phys.org/news/2016-03-seeding-iron-pacific-carbon-air.html

29 https://nap.nationalacademies.org/cart/download.cgi?record_id=26278&file=77-102

30 Huge thanks to Taylor Mattie, Kate Murphy, Ken Buesseler, Andrew Cantino, Shantanu Agarwal, Jamie Beard and Kumar Garg for providing comments on earlier drafts of this essay that made it much better and helped us avoid saying a variety of incomplete (or stupid) things.