Seemingly everyone in carbon removal says they want more data transparency and the sharing of scientific results. Why isn't open science more present, and how can we get more of it? Could a pre-print server for CDR be part of the solution?
Today is the official launch of CDRXIV ("cee-dee-archive"), a new initiative from CarbonPlan that aims to spur scientific conversations within the carbon removal community.
On this episode, Freya Chay (the CDR Program Lead at CarbonPlan and a Member of the Advisory Board to CDRXIV) and Tyler Kukla (a CDR Research Scientist at CarbonPlan and the Content Manager for CDRXIV) are on the show to explain how pre-print servers drive progress in other scientific fields, why CDR needs one, and how it may change our industry.
If you'd like to submit data and/or a paper for publication to CDRXIV, please email hello@cdrxiv.org, or visit their submission portal here.
This Episode's Sponsor
Listen to the RCC episode with Lisett Luik from Arbonics
Become a sponsor by emailing carbon.removal.strategies[at]gmail.com
Use this affiliate link to use Descript's transcripting and podcast editing service
Use this affiliate link to use Riverside to record your podcasts
Resources
Become a paid subscriber of Reversing Climate Change
Full Transcript
Ross Kenyon: Before we begin, I’d like to share a few words from our sponsor, Arbonics. Thank you, Arbonics, for sponsoring Reversing Climate Change. You helped make this show possible. I’m so grateful to you, and you deserve it. I don’t like taking sponsorship from people whose companies I just don’t believe in — I don’t want to work with them. And thankfully I don’t have to, because there are companies like Arbonics who are willing to come and support Reversing Climate Change.
If you don’t know about them, they’re doing amazing forestry projects in the Baltic states in Europe. Europe has over 14 million hectares of underused land. A lot of it’s abandoned or low quality. Not a lot is happening on it. It used to be forest, but it was cleared for farming as farming intensified, and then a lot of that farmland was later abandoned as the global food system moved towards lower cost producers. The land just wasn’t worth tending, and certainly not tending in a way that we would respect as a regenerative and carbon sequestering process. It was just underutilized, maybe unloved.
Arbonics uses technology to help landowners even find this land, and once they do have it, how to restore it back into carbon removing, biodiverse forest. If you care about carbon being removed and you want to see Europe return to the forested continent that it once was, you should be looking at Arbonics. Go talk to them. I had Lisett Luik, their founder, on the podcast. The link is in the show notes if you’d like to listen to it. We talk about a lot of the tricky issues that surround durability and carbon removal, and how does forestry fit in. It’s a very thoughtful episode. I really like talking to her.
They made it really easy to pitch Arbonics on doing some business together, because I want to see their work be successful — restoring biodiversity and making forestry, especially good forestry, profitable again. Man, that is a game changer, and so important if you can pull it off. Thank you again, Arbonics. The link to check out Arbonics’ website is in the show notes. You should listen to the show if you haven’t heard it already that I do with Lisett. It will teach you quite a lot about how they think and what they’re doing. And now we will fade back in to the rest of the show. Thank you for listening. Here it is.
Hello and welcome to the Reversing Climate Change podcast. I’m Ross Kenyon. I’ve been involved in carbon removal for the better part of a decade. I’m a climate tech entrepreneur and I’m really excited about today’s show. Before I introduce the show, I want to take a moment to thank everyone who has become a subscriber so far. Your support is amazing to me. It really is incredible that so many people want to contribute $5 a month to help make sure that the show continues. The link is in the show notes. It’s done through Spotify. It’s very easy. It gets you ad-free listening, bonus content, and my very sincere gratitude.
So please do that if you haven’t yet become a subscriber. Give a great rating and review on Apple Podcasts and/or a five star rating on Spotify. That helps a lot too, and is also very much appreciated. And now I’m going to tell you about today’s show. If you’re familiar with carbon removal, you’ve almost certainly encountered the work of CarbonPlan. CarbonPlan puts out an enormous amount of research on various topics, much of which relate to carbon markets, and carbon removal is very data-driven. Groups such as CarbonPlan serve an extremely important role within carbon markets. Having watchdog organizations make sure that what is being claimed is actually happening is incredibly important. I really admire what CarbonPlan is trying to do right now.
CarbonPlan built and maintains CDRXIV, which I think is a great thing for them to do. CDRXIV is a preprint server that allows for material that will ultimately go through peer review — or much of it will — but can be preprinted without review basically immediately. This is great, because peer review can take quite a long time, and this is an area where we really don’t want to spend huge amounts of time waiting for results to be confirmed before even discussing papers. Obviously things need to go through peer review and make sure that this work has been done appropriately and reviewed with the correct amount of scrutiny, but in order for the conversations to continue advancing at pace, preprint servers do very important work in a number of other disciplines.
That CarbonPlan is now bringing to carbon removal. And carbon removal is very famously interdisciplinary. There are so many types of scientists that I think before carbon removal probably weren’t coming into contact with each other very often. In fact, someone told me recently — and this was a geochemist who told me this — that geochemistry was a backwater until CDR plucked it out of obscurity. It’s just now having its time in the sun, you might say. I really like work that’s trying to break us out of competition. People always talk about how we’re too competitive and people need to open up and share more information and do more learning in public. And it’s one of those things where I hear a lot about it, I don’t see as much of it as I might like.
And this is a great illustration of why nonprofits can be so valuable in for-profit spaces: they can make it much safer, at least partially, for those who work at commercial companies to go to their commercial colleagues and say, hey, I’m doing this research, we should put our data that we actually have a proprietary right to — we do not have to share this, but we can put it up on this preprint server called CDRXIV that CarbonPlan runs — and that will allow us a forum to have discussions about our work that brings legitimacy to what we’re trying to do and shows leadership, that we are trying to do this work in public so all of our peers can learn.
It’s good for all of us when any one of our quote unquote rivals succeeds. There is still a great deal of growth, I suspect, for so much of carbon removal, that even those working within the same niche — I think there’s plenty of room for everyone to go after and still have a pretty nice chunk of the pie. And I think efforts like CDRXIV have the right values to make that legible to those wishing to publish data, and to give them a forum to convince their colleagues. If anyone wants to reach out to CDRXIV, wants to get involved: hello@cdrxiv.org. They have a clever way of writing archive there. The link is in the show notes. Thanks for listening. I hope you enjoy the show.
The guests that I have today are both from CarbonPlan and both work on CDRXIV. Freya Chay is the carbon removal program lead at CarbonPlan and she is on the advisory board of CDRXIV, and Tyler Kukla is a research scientist of CDR at CarbonPlan and the content manager at CDRXIV. Thank you, Freya and Tyler, for being on the show. I’m really glad that you’re here, honestly. Thank you for being a voice for open science. I hope others will hear what you’re doing and emulate you. Get in touch. Thanks so much for listening. That’s enough prologue. Here is your show.
Tyler and Freya, thank you for being here with me.
Tyler Kukla: Thanks for having us, Ross.
Freya Chay: Thank you.
Ross Kenyon: Tyler, does this feel weird that we don’t have beers right now?
Tyler Kukla: It does feel a little bit. I am missing our Seattle-area mCDR happy hour. It’s also weird that we’re in Seattle and we’re not together, but I think that we’re going to make it work.
Ross Kenyon: I think we’re going to make it work too. I’m happy we’re doing this show. We’ve been talking recently about open science and how important it is, and also how absent it is from so much of carbon removal. CarbonPlan has a new initiative called CDRXIV, which I’m not sure I would have guessed from how you have it written out, but it is a very clever way of styling this new initiative. Would one of you introduce CDRXIV?
Freya Chay: Sure. Yeah, CDRXIV is a preprint server, so it’s a place to share early research and data on carbon removal. It’s modeled off of other archives. There’s a physics archive and a bio archive. Essentially, they are infrastructure for open science values. Basically, how do we do science in a way that is transparent, that is accessible to everybody, and that is fast — that really enables, I think [unclear]’s tagline is, communication to happen at the speed of science.
Ross Kenyon: What does it look like when science is done without a preprint server like this? What does science even look like?
Tyler Kukla: Something from the academic side: should you finish some kind of project that you’ve been working on — often for, you know, a couple of years; it can take shorter than that, but also it’s a long process — you want your results to be made publicly available, or at least available to the readership that you care about. And you write it up as a paper, and you submit it to some academic journal. Then it goes out for peer review. The peer review process can take anywhere from three-ish weeks all the way up to a couple of years, depending on how many iterations happen, depending on how responsive the peer reviewers are, and often the culture of the academic journal itself.
And what that means is that something that you discovered, or some new progress or new step forward that you have made in the lab or in your models or in any discipline, can take a very, very long time from the moment that you make that new result and that step forward to the moment that other people have access to that information. And the process of peer review taking, you know, up to a year or more is a real challenge when you want science to be moving quickly, and when science itself is moving a lot faster than the stuff that actually ends up on your feed of new results that are making it in the published literature.
Ross Kenyon: Do journals like services like these?
Freya Chay: You should ask them. I mean, I think the culture around preprinting varies by discipline. So in some disciplines like physics, most of the activity is really around arXiv and preprints — sharing preprints, reading preprints. That’s where people are really tracking progress. In a discipline like biology, there is growing activity around bioRxiv. A lot of that work will eventually also go through the peer review process and end up in a journal. So bioRxiv is a place where you can see those results early, disseminate results early, foster a faster conversation between an intellectual community. But it’s not replacing the peer review process, it’s just speeding up the conversation.
I think a really great example of that was during COVID, where you had, for instance, the COVID genome ending up on bioRxiv within a couple of days of it being sequenced. Like, that was a conversation that needed to happen fast. There was a lot of academic follow up to that work, but the preprint server was really a place for a quick conversation among an intellectual community to happen, that then went on to go through the peer review process and show up in journals.
Ross Kenyon: Does preprinting ever have a dynamic where sensational articles get more attention than they would if they had been through the peer review process?
Freya Chay: It’s a really good question, and I think it actually is best answered by maybe taking a step back and talking about the culture of open science. So preprinting is one tool for open science, but if you really step back, the value that is underpinning preprinting is that science should be transparent, it should be inclusive, and it should be effective. When you really cultivate a culture of open science, you end up using something like preprint servers, but it comes alongside the recognition that it’s your responsibility as someone consuming that material to bring your own judgement about what a valuable contribution is.
You’re not relying on a journal stamp of approval. Something’s not more valuable because it ends up in Nature. It’s really a conversation among peers. So yes, it does require more discernment. And I also think that’s actually a really constructive dimension of the culture of open science that is supported by other practices.
Tyler Kukla: Yeah, I would add that there’s two sides of that coin. So on the one hand, the things that end up on the preprint server, usually by the time that you’re looking at them, it hasn’t gone through peer review yet. You can of course update the next version of your manuscript to a preprint server after it has been around, or after peer review. So that doesn’t mean that everything that you read inherently hasn’t seen the touch of peer review. But on the flip side of that coin, a lot of the feedback that we’ve heard from folks who have been using preprints for many, many years — they’ll tell us that, look, some of the best feedback that I get on the results that I make available happen because I put it on a preprint server.
The peer review process often engages two to four, if you’re lucky or unlucky, other peer reviewers to take a look at the research that you did. And as research becomes more and more interdisciplinary, it’s hard for those two to four people to have the expertise to span all of the stuff that you did in your paper. Which means that there is definitely stuff that slips kind of through the cracks through the peer review process, because nobody that looked at it in the process of peer review might have the specific expertise needed to assess something that you did.
But if you make something available as a preprint for everybody to look at, you have a better chance of catching all of the relevant people along the way. And that tends to lead to better outcomes. One of the things we want to encourage is more of this feedback and communication around these preprints, because that’s what you really need for that mechanism of a more comprehensive set of feedback to work.
Ross Kenyon: Does one reviewer always have to be mean, or is that not the case? I know that’s the reputation.
Tyler Kukla: Yeah, it’s usually — we say it’s Reviewer 2, or whoever was the last reviewer. They get numbered like 1, 2, 3, 4. The last one to turn it in is the last number. And then usually the last one to turn it in, it’s because they really didn’t want to read the paper that they got sent. And so, no, you do not need to have a mean reviewer. And I wish that peer review was a nicer process, but it can have these mean elements. And it’s usually that last reviewer that you want to be emotionally prepared to read.
Ross Kenyon: I think the benefit here is pretty clear, why we would want to engage in science in this more rapid way, even though peer review is clearly a very important part of empiricism and the scientific method. We’re talking about open science, and I think people could probably infer parts of what you’re saying as being constitutive of open science, and maybe we should define it clearly and contrast it against what might closed science look like.
Freya Chay: Yeah, that sounds great. So as I said before, I think open science is really rooted in some values and in the culture. And in my mind, the easiest way to talk about that culture is kind of expanding your understanding of who is on your team if you are a researcher. So I think that really starts with the process of research. What does it look like to create collaborations in which people can really understand and participate in what you’re doing? What tools do you reach for if that’s your goal? What kind of documentation do you make around your methods?
Then there’s this like next layer out, which is how do you share your findings? How do you share your findings early, quickly, and in the open? How do you share them in a way that people can give you really good feedback? How do you not just share your findings, but share the data and the code and the tools you built with permissive licensing that lets other people on your scientific team — which is this whole community — use them and build upon all the work you’ve already put in? And how do you be transparent both about your methods and about things like conflicts of interest? I think all of those practices underpin this culture of really seeing science as a team effort that’s in the public interest.
Ross Kenyon: How can a broadly commercial community like carbon removal — maybe you disagree with that characterization, but it certainly has a number of commercial interests that interweave through the process of science. Is carbon removal an example of an open science community? Are parts of it openly scientific? Where is it and where is it not? You can take it any way you want.
Freya Chay: Yeah, it’s a great question. So I think it’s interesting to talk about the academic community working on carbon removal and the kind of industry or industry-adjacent community as, you know, very interrelated but maybe separate states. So I think in the academic community, it’s not a norm to be participating in open science. I’m not sure there is really yet a single CDR kind of like topic area academic community. Instead you have kind of still disciplinary communities who are thinking about particular types of CDR. You certainly have certain PIs or labs who are very committed to practices like preprinting, but CDR as a topic area in the academic world, I don’t think has the type of culture you might see, for instance, in physics where everyone is using preprint servers. A lot of CDR research will just end up in journals. So I think there’s room for growth there.
But examples of leadership certainly from some labs. In the kind of industry and industry-adjacent NGO funding space, I think we have so much room for growth. Again, I think you have examples of people pushing in the right direction. We can go into those — people who are trying to share data from industry-led field trials, who are either doing those field trials or trying to facilitate the data sharing. You have examples of funders who are starting to ask for data sharing. But if you just think about how much we are learning through industry efforts in the field, and what fraction of that is publicly available to anyone to build upon and learn from, I think we have a pretty big and clear area for growth.
Ross Kenyon: How do you feel about all of that, Tyler?
Tyler Kukla: I think that feels right. And I think that it’s worth thinking about, like, okay, there’s this room for growth, but what is the prize at the end? And when you make more information openly and publicly available, I think that there are a bunch of great reasons to do it. And one of them is that when we talk to folks in the CDR community, it’s pretty clear that a lot of folks are in agreement — pretty much everybody’s in agreement — that we want to figure out what works and do more of the stuff that works. And we want to do it quickly. We want to do it as quickly as we can to keep up with the pace of climate change, really.
And open science is a way to do that. It helps us get over the things that don’t work faster. It makes the mistakes and the missteps just little hiccups along the way, rather than big entrenched problems that grow into something that can erode or create challenges down the line. It serves a lot of the purposes of what we’re trying to accomplish as a CDR community. And I think that that spans both academia and industry. And I think it’s worth keeping that in mind as we’re talking about what is the right way to build an open science culture in CDR. Ross, I feel like you’ve thought about this too. What’s your answer to that question?
Ross Kenyon: I can tell you a funny anecdote about it. Maybe this will partially answer it. I’ve had so many conversations with project developers or people that are working on the commercial side: yeah, why aren’t we more transparent about pricing or about the science that we’re doing internally? I’m like, yeah, why don’t you just publish that information? You could just do it unilaterally. Like, well, not alone — like, okay, I can’t just do that all by myself. What if no one else does it? But clearly then you just have, you know, bad prisoner’s dilemma dynamics. The game theory of it prevents anyone from doing anything. And we end up in this suboptimal equilibrium here where none of us learn and it’s all private.
I don’t even think that the privacy is that valuable either. Everyone is so focused on IP, and that’s often driven by venture capital dynamics. But I don’t think a lot of this IP is going to be, in the case of wind down, super valuable. I don’t think that we need to be as focused on patenting as we are. I don’t think that being secretive is that valuable of a moat either. Tito Jankowski of AirMiners has a line that I’ve heard him say so many times, which is that our chief rival is obscurity, you know, not one another. It’s more likely that carbon removal will just fade out into nothingness.
And so what? Like, if you’re in enhanced rock weathering and Mati wins the grand prize of XPRIZE, are you really worse off? In fact, you probably raise the profile of everyone in the entire space by that happening. But yeah, I think we end up being competitive in ways that are actually not even smart if you’re a Machiavellian. They’re just sort of suboptimal. Even if you’re a ruthless, self-interested person, it’s still bad. So I’ve always tried to lean that way and encourage people to go that direction. I think unilateral action can look like great leadership and be commercially valuable for companies to do. Yet no one takes my advice on it. I don’t know why.
Freya Chay: Yeah. I mean, that all really resonates with me. I think there’s such a strong argument that the nature of what is trying to be sold in this industry has such intangibility and such scientific uncertainty that the benefits of the potential learning and making it trustworthy and real so far outweigh the potential losses to a business model. But I have also never run a startup, so I recognize that I have a limited view here.
Ross Kenyon: One of the resources I shared ahead of doing this podcast was an episode that we recorded probably at this point about a year and a half ago, internal to Nori, about the direct air capture and storage methodology that we created. And we decided at that time to publish our methodological work — of which there were three — under a very permissive Creative Commons regime that allowed others to remix our work, to reuse it exactly as written. Other registries, Isometric or Riverse, could just pick it up and issue credits through it, should they like it enough. And I was hoping to see more of that, but unfortunately, as far as I know, no one has taken that up and run with it.
And I’m not even sure it’s that great of an area to compete against one another on, too. Like, does it make a lot of sense for Puro and Isometric to be neck and neck on methodological development? Is that actually in the interest of our industry? I imagine you’re probably going to say no, but maybe you’ll surprise me.
Freya Chay: I don’t think I’m going to surprise you on this one. Yeah, I mean, I think historically when we’ve seen competition on methodology development, it’s created a race to the bottom. How do you make a methodology that you can still sell a credit from, but it’s easier for folks to comply with? And a certain amount of that I think is useful. We don’t want to create methodologies that are unusable — that’s an important thing to be attuned to. But yeah, if the story we’re telling about methodologies is that they accurately help us quantify real world outcomes, then the idea that we should be sharing the ability to get the best understanding of real world outcomes with each other feels extremely obvious.
And I might actually extend your observation about registry licensing of protocols to also the data that’s on registries. We recently built a tool at CarbonPlan called OffsetsDB — offsets database. And it scrapes a bunch of project data from all of the big traditional voluntary carbon market registries. And one of the really tricky parts of that project was navigating the licensing around data. So yeah, you can go on their website and freely use it, but a lot of that data is actually licensed in a surprisingly restrictive way that makes it legally questionable to repackage it, make it public, do work on it. And that seems to me like a pretty big problem if we’re talking about registries being a mechanism for transparency in this field.
Ross Kenyon: Wow, talking about this just makes me feel like — am I actually a climate hippie and I’ve been denying it this whole time? I thought I was a hard-edged businessman. Turns out I’m just like, data, it needs to be free.
Freya Chay: Yeah, yeah. Come, come to the nonprofit side, Ross.
Ross Kenyon: CarbonPlan has a reputation as being a very strong and powerful watchdog, and trying to make sure you end up on the right side of the science and are attuned to quality standards in a way that allows you to claim that you’re producing credible carbon removal is actually genuinely difficult. And so Tyler and I spend a lot of time talking about fundamental questions of enhanced weathering that are still up in the air in some ways. And yet there’s commercial developers who are basically assuming that it works and broadly works the way that it is intended to, even though there are so many open questions here. And the tension is often between how do you run a business that is reliant upon science when so much science is still up for grabs and still being worked out?
I think even if you have good intentions and you mean well, you can still face incentives that are really difficult for commercializing in the face of this uncertainty. The tension and the friction between scientific integrity and commercialization, it runs really, really deep here. Does it create a fundamental conflict of interest? What does it mean to be on the non-commercial side of this, watching commercial developers do things with science that you care a lot about, that maybe you don’t feel great about sometimes?
Tyler Kukla: I think one thing that’s interesting is, in talking to a lot of the researchers that are involved in these startups, I think that their read on the literature tends to be pretty similar to mine, which is that we still have questions that we’re trying to answer. And also that our startups that have all this wonderful funding and infrastructure to build out monitoring systems and so on and so forth are a really great vehicle for starting to try to answer some of these questions.
And the challenge that we are faced with then is like, okay, you might have orders of magnitude more data than I’m able to access through what’s available in the academic literature. And in fact, that data is probably a lot more useful for thinking about what an enhanced weathering deployment might look like in the real world, because academic studies aren’t subject to the same constraints as a company trying to operate and do their own field study, where we’re, you know, working with spherical cows often in kind of academic land. And on some spectrum, some academic studies are a lot more close to what a real world deployment looks like than others.
But at the end of the day, the question is, okay, if you’ve generated new knowledge by collecting way more data than anybody else has access to, do you have a responsibility to share that new knowledge publicly to help pull the rest of the field forward? And is it possible that you can just share that new knowledge with funders and convince them that this is all buttoned up? And do we have this world where there are people that are more compelled or convinced about the state of knowledge in enhanced rock weathering, because the fact that some data is kept under closed doors and shared internally, than the rest of the public is? And I think that those are hard questions to answer.
But I do think that one of the prizes of open science is that you start to get data synthesis. You can bring a bunch of different data sets together to paint this picture that nobody else can see from the specific experiments that they’re working on or the point sources that they’re looking at. And this is where we often see across fields like big steps forward in scientific progress and understanding. And I think that if we really think — as I think people agree — that there are barriers to understanding right now, there are knowledge gaps, the potential for big data synthesis like this, by making those data available, can I think do a whole lot of great work at plugging those knowledge gaps up. The challenge is how to do it, of course, but I do think it’s challenging to walk that line around how much data is being produced and how do we make that data useful for public learning at the same time.
Freya Chay: I’m going to add, in response to your question, Ross, one of the things that really strikes me as a piece of this puzzle that makes things really hard is the offsetting claim that comes at the very, very end of the chain. And if we were operating in a world where we weren’t trying to make claims about physically undoing the harms of a fossil emission, the leeway for uncertainty and the consequences of being wrong at any given time step, they’re dramatically lower.
So part of my curiosity is like, wow, in what other world — like what other levers do we have, policy levers, private sector levers, to allow learning by deployment without the pressure of the offsetting claim at the very end? And this is a curiosity I’ve had for a long time, and I think it matters for all sorts of things like soil carbon and forest carbon and uncertain long duration CDR. Not an easy answer, but I think it will be really interesting to see how that evolves over the next five years.
Ross Kenyon: That’s such an uncomfortable space, though, where so many people in our sector come from STEM environments where there often is some amount of objectivity or presumed objectivity, and you don’t want the questions to end up in, well, it ultimately comes down to political economy yet again, where we’re telling stories about what these credits are good for and how does that impact what corporations and governments are allowed to do in the meantime. The amount of indeterminacy in a solution like that, I think, makes a lot of people uncomfortable. That’s a really hard question. Unfortunately, it may just be the truth.
Freya Chay: Yeah, yeah. I guess my lighthouse is always like, we’ve got to learn how do we tell the truth about what we know all along the way. Because I think as soon as you start overstating what we know — which I think offsets ask us to do for a lot of these technologies, if they’re the only funding vehicle — then I think we’re really counter to open science values. We’re creating this obscurity that really doesn’t allow the full community to seek truth together. And I think that’s a really counterproductive dynamic over the mid to long term.
Ross Kenyon: Do you have some alternative in mind relative to the offsetting paradigm?
Freya Chay: I mean, if I had the perfect solution, I would have tried to write about it or convince someone to do it by now. I mean, there have been lots of experiments. So I think there are experiments where you’re still fundamentally thinking about carbon credits, but you’re using an expenditure model and not supporting offsetting claims. So that’s one kind of pressure relief mechanism. I think the whole conversation around beyond value chain mitigation — imperfect, but another example of someone trying to release that pressure valve.
I think there’s a blossoming conversation right now about policy mechanisms that could support CDR activities in a way that doesn’t require offsetting claims made by the government or made by regulated private entities. I think all of those are really, really interesting. There’s not a clear answer yet, but worth continuing creativity, I think, especially on the policy front when we’re thinking about how we invest public resources in developing high potential but deeply uncertain carbon removal approaches.
Ross Kenyon: There’s some line — it’s either told about A Brief History of Time or The Elegant Universe, one of those pop physics books — about how for every equation that was added to the book, sales would drop an order of magnitude. And that always makes me laugh. But Nori, when we were active, we only ever felt comfortable guaranteeing 10 years of carbon storage in soils. And that was such a hard story to explain to buyers who’ve been coached since Kyoto that they just buy a ton and they don’t have to think about it again. And instead we have to explain different types of stacking, and why vertical’s not really considered that great, but horizontal maybe.
And you just lose people because it’s really complicated. They just want to, like, what do I pay? And then can I move on to the next thing of 100 that is on my to-do list? And this is part of the tension here of doing good science and having rigor versus just selling the thing that moves the ball along. And I’m not sure that we ever figured out a great solution to it, because it bedeviled us and was a really hard problem to solve. Customer education, making sure they actually understand what a credit is good for.
Freya Chay: Yeah, it’s tremendously hard. The real world has a lot of details, and details are hard to organize ourselves around.
Ross Kenyon: I hear that.
Tyler Kukla: I think another challenge that emerges is that even if you made all the data that companies are collecting publicly available, that still might not get you exactly as far as you want to go in terms of feeling really good about how effective a given removal method is. You know, companies in enhanced rock weathering, for example, are collecting soil samples that are — I don’t know how deep they are, because we don’t have the data, but I have to guess something around 30 centimeters or so, maybe 50 centimeters deep on the high end. And that’s a very small fraction of the total distance that carbon is going to travel before it ends up in some durable storage reservoir.
And so we have this other question around how do we support not only just making data available so that we can drive and accelerate learning, but how do we develop mechanisms for doing the work that also directly support the learning process? How do we develop mechanisms for doing the work that help us sample places that we might not be sampling otherwise, if all that we need to do is demonstrate that something happened in the top 30 centimeters of the soil profile? And I think that that’s a hard thing to do. Like, funding that supports learning is difficult kind of across all sorts of different funding models.
Ross Kenyon: If you could guarantee compliance with your wishes, what would you design so that this was no longer a problem?
Freya Chay: Oh, can you say one click more?
Ross Kenyon: Well, just assuming that you don’t have to face the game theoretical issues of people having to opt into this. What if we just knew they were committed to doing the right thing?
Freya Chay: What do you mean is the right thing?
Ross Kenyon: Is it as simple as just publishing all of this data all of the time and giving you access to it so experts can parse it in the open?
Freya Chay: Yeah, it’s a great question. Okay. So we’re saying basically, assuming business models didn’t create big barriers to participating in open science, what would it look like for CDR companies to really implement those values in practice?
Ross Kenyon: Yeah, it’s a climate hippie question.
Freya Chay: Climate hippie question. Yeah, I mean, I do think one piece of the puzzle is really thinking about how to share data and tools publicly and quickly. I think part of the mindset of open science is that you don’t know who your findings will be useful for and you don’t know when they will be useful. And the idea that you are creating more surface area for collaboration and for unexpected insight by sharing stuff, even if you don’t have a super clear story about who exactly is going to use it for what — that’s really core to the open science project.
So yeah, if I was waving a magic wand — and I do think there are important caveats around data quality, data interpretability, data provenance — but I think we would be seeing a lot more data sharing with no restrictions. So CDRXIV is one place that you can put a data set, you can get a DOI, it can be clearly linked to you. If someone uses it, they have to give you credit. But there are no restrictions. Anyone with internet access can see it, they can ask questions about it, and they can reuse it. That kind of data sharing I would love to see a lot more of, and I would love to see that not only for positive results, but for null results.
So I think this is another kind of deep tenet of traditional academia that’s tough: that you get credit and you get shiny papers when you publish exciting results, and null results are super important to figuring out what doesn’t work, which is a huge part of the CDR project. But they don’t get you papers in academia and they don’t get you credits in carbon market land. So what does it look like for startups to be sharing more information about what didn’t work with each other and with the science community?
And then I guess the last thing I would say is right now, even if we’re not in that perfect world where you’ve, you know, sidestepped all of the incentives that make sharing hard — transparency about what exactly is making sharing hard. I mean, I think the collective action problem in general is floating around and I understand it’s really tough. But beyond that, we have lots of general conversations about transparency. How do we make it more specific? How do we disclose when you choose not to share, exactly why, in ways that help other people problem solve around that as a community? Those would be kind of my three top ideas. Anything to add, Tyler?
Tyler Kukla: Yeah, I agree with all of that. And I would just add that if we live in a world where you can just throw all of them in there and make it publicly available, I think it would be better. I think that sounds great. And also what you really still need is that culture of open science. Like, a lot of what makes open science work happens before anything actually is made publicly available. It happens in the culture that you develop in your lab, the tools that you use to share your files across all of your collaborators to make sure that everybody sees what you’re up to and can contribute and pitch in in ways that, as Freya said, might end up being a little unexpected.
So that you have all of that infrastructure in place so that you can make your information publicly available when the time is right, when the project has concluded, or whenever that is. In order for open science to really work, you can’t just throw a bunch of stuff at the wall and hope that everything is going to stick or find a place where it can add more value than it otherwise might if it just lived and died with the thing that you said about it. You need to build infrastructure around that as well. And that’s why we think about open science as a culture, because when you think about the culture of open science, it implies that you have some additional building of infrastructure and nourishing of the broader system that goes along with it.
And so that makes it easier to, you know, if you are making all of your data publicly available, the infrastructure around it makes it easy to direct your attention and find the stuff that you’re looking for and make sure that it’s not an inundation of information, but it’s a totally navigable sea of information. And that is a tough thing to build, but it does require that culture. It takes time. There are a lot of great tools to do that already, and we see CDRXIV in large part as a place for that — like, one example of infrastructure to support open science to take place. And I think there’s a lot more that we can do.
Ross Kenyon: If you have someone’s attention who is a project developer working in carbon removal, how might you convince them? Right now we can conclude with this: make your plea to the carbon removal community.
Tyler Kukla: One thing that I’ll say is that coming from an academic world, when we talk about open science, it’s almost entirely talked about as like a grassroots thing. And it should be — it is a cultural thing, as I mentioned. But in large part of academia, we talked about it as, this is a grassroots driven movement, in part because the institutions are so slow to respond. You need to build a lot of momentum before you get top down support for the resources that are needed to make open science work.
And I think that we’re lucky that that’s not really the world that we live in in CDR. A lot of the top down institutions and buyers and funders are really dynamic and they’re keeping up with the science and they’re trying to understand what is going to help research and knowledge and understanding progress quickly. And when you have those elements of the system, I think it’s really easy for the top down institutions to not be followers of a momentum, but actually be leaders and push things forward.
And so I don’t know exactly what that’s going to look like, and I would love to see, you know, more requirements around data sharing as one example. But I think it’s just important to note that we don’t have to rely on institutions that are moving slowly here. I think that the top end institutions are really dynamic and flexible and thinking hard about this. And so I would place a call to action there, as an opportunity to lead a shift toward a more open science culture in the community.
Freya Chay: Yeah, plus one. I think what I would add for a project developer is, I really think this project of carbon removal is only going to work at scale and fast enough if we expand our idea of who’s on our team, and if we work together, and if we learn a lot. And we recognize that not all the incentives are there and not all the infrastructure is there. But you can lead by example. You can help start breaking the collective action problem. And I think you will get credit for that. I think you will be respected for that, and you will contribute to the type of collaboration and external examination that ultimately I believe will give you a path to, you know, long term project development in a sustainable and trustworthy way.
So I recognize it’s not easy, but this is one opportunity. CDRXIV is one opportunity. There are other ways to embrace those values. And I just really encourage people to lead by example and start conversations about the specifics of, like, how do we translate these open science values and the potential benefits they bring into our industry in a more granular way?
Ross Kenyon: If someone is writing a scientific paper that may be applicable to the server, how do they get involved with CDRXIV? How do they submit? What’s on there now? What’s the status of it?
Freya Chay: So right now I think there are 13 preprints on CDRXIV. The process is really simple. So you can show up on CDRXIV, you make an account, and there’s a pretty simple submission process where you can upload either an academic style preprint, a preprint plus data, or just a data submission. That submission goes through a very light touch review process, which relies on a really awesome set of external scientists who really are just checking to make sure that a new research result is presented. It’s not a peer review, it’s just a light touch check to make sure it’s about carbon removal and there’s actual research there. And we commit to communicating back to you within three days.
And assuming that submission meets CDRXIV’s criteria, it will be posted publicly with a DOI on the website. And after that you have the chance to update or change your submission, and there’s a version control system. But once it’s on the website, it will remain publicly available associated with your DOI forever, no matter where that work ultimately ends up, or if it ends up on a paywall. Anyone with internet access will be able to read it and learn from it.
Tyler Kukla: I was just going to add one more thing, which is that in talking about open science in general, CDR is similar in that it really works when people are a part of that community, when people are contributing results, and when people are going there to look for new results. We think CDRXIV makes sense because CDR is a topic that covers a crazy number of disciplines — like, so, so many disciplines — and the existing platforms for sharing new knowledge are disciplinary. And this makes it so new knowledge gets spread all over the place. And it’s really useful, if you want to be learning about CDR across those disciplinary boundaries, to have things organized under a topic.
We also recognize that not everybody is doing work in CDR that is going to end up in the peer reviewed literature, and not everybody is taking the time to write long format articles. And so one of the things that makes CDRXIV different from other preprint servers is that we also have a data repository feature. If you’re producing a data set but aren’t maybe going to spend the time to write a paper around it, or it’s data that didn’t fit for a paper or whatever it is, you can submit the data to CDRXIV and it can be discoverable as a data submission by itself. And this is one of the ways that we’re trying to make sure that we are catering to the broader CDR community and not just to the people that are writing articles and papers.
And it’s not the end. We have a lot more things that are kind of on the horizon that we want to think about. If there are ways that you think that CDRXIV can help cater to the type of research that you’re doing and help make your work publicly available, we would love to hear from you. We’d love to hear about it. Ross, maybe I can give you an email address to put in the show notes or something.
Ross Kenyon: Yeah, that sounds good. Links to all of the things that we discuss are in the show notes if you’d like to follow up and get involved. Freya and Tyler, thank you for being here. Very important topic. Respect very much the work that you’re trying to do, and thanks for doing it, and thanks for being here.
Freya Chay: Thanks for having us.
