Scientists and engineers around the world have developed dozens of approaches for carbon dioxide removal (CDR), including such oddly-named approaches as “biomass slurry injection” and “coastal enhanced weathering.” The naming conventions for these approaches are often independently developed, and because there are so many, it can be difficult to keep track of their similarities and differences. To help, the field has produced various taxonomies that provide groupings and sub-groupings.

Taxonomies are useful for seeing similarities and differences across approaches, but by linking certain approaches, they also restrict our ability to see other, alternative groupings. Here, we propose a new way of understanding the CDR ecosystem that steps outside of existing CDR taxonomies and instead looks at the flow of a carbon atom through its removal lifecycle. We believe that, by stepping outside of the conventional taxonomic mindset, this method of thinking has the potential to unlock scientific breakthroughs, commercial synergies, and non-technical barriers.

CDR taxonomies

Since its origins, the field of CDR has produced various taxonomies, each of which has value for different use cases. For example, RMI developed a taxonomy with primary categories based on the “key input” to the approach — biomass, alkaline materials, or low-carbon energy — which we refer to as biogenic, geochemical, or synthetic CDR. This taxonomic approach is useful for understanding input-based limitations to the scaling potential of CDR.

Another common taxonomy, originally proposed by XPRIZE, includes the categories of Air, Ocean, Land, and Rock CDR. Because it groups CDR approaches by process-related similarities, it is helpful for understanding and navigating the social, legal, and behavioral challenges of CDR deployment. For example, all Ocean CDR approaches must grapple with the physical, ecological, and social aspects of interacting with the ocean, including tides, marine life, and ocean communities. Our recently published roadmap, co-authored with the Bezos Earth Fund, used this taxonomy to discuss not just research, development, and demonstration (RD&D) milestones for different categories, but also necessary activities related to community engagement, policy, and market development.

Because these taxonomies parse the same CDR ecosystem from different angles, they have significant overlaps. For example, Ocean CDR approaches may require key inputs of biomass, alkaline materials, or energy, but they all have a common challenge of interacting responsibly with oceans and ocean communities. Likewise, Geochemical CDR includes Rock CDR approaches and can also be another way to describe some Ocean and Land CDR approaches.

A new approach

Another way to make sense of the wide range of CDR approaches is to follow the fate of a carbon atom on its removal journey as it flows through three major steps:

1. Capture: The capture step is where carbon dioxide is removed from the atmosphere
2. Process: The process step is how the carbon is modified into a form compatible with storage
3. Storage: The storage step is the final resting place of the carbon atom

At each of these three steps, several things can happen. For example, at the capture step, carbon dioxide can be captured by reaction (such as photosynthesis), physical forcing (pressure/membrane), passive water contact (absorption), or sorbents (adsorption). The captured carbon dioxide can then be processed in a variety of ways, such as chemical transformation (heat/microbes/combustion), mechanical processing (direct air capture/direct ocean capture), or ecosystem management to preserve captured carbon in forests, soils, or aquatic environments. Finally, the captured carbon can be stored in an organic form (stabilized biomass), an inorganic form (mineralized carbon), or as stored carbon dioxide. It is also possible that some carbon will be released through losses that can happen at various points along the way. For example, during their processing step, pyrolysis-based approaches re-release up to 50 percent of the carbon that was originally contained in the biomass.

Notably, this flow chart maintains conservation of matter across the 3 steps, so the carbon that enters the diagram on the left cannot be destroyed or lost. If it enters on the left, it must in some way exit the diagram on the right. For this reason, the total amount of carbon at each of the 3 steps is the same. Furthermore, the carbon cannot flow backward.

When placed on the flowchart, it is now possible to see that individual CDR approaches (we used the naming conventions and numberings from our Applied Innovation Roadmap) are based on different stages of their carbon as it moves across the CDR flow diagram. Some are named after their capture step (e.g. microalgae or macroalgae), some are named after their process step (e.g. biochar), and some are named after their storage step (e.g. biomass direct storage). The flowchart allows us to see which steps are shared between different CDR approaches, suggesting overlaps in scaling challenges such as resource requirements or policy needs.

In biochar, for example, carbon is captured in plant-based matter, processed in a pyrolysis unit, and then stored as biochar. With the flowchart, we can see that the biochar pathway is defined by its process step, not its capture step or its storage step. So, it might be possible to use both different capture steps (such as marine biomass), as well as a different storage step (such as sinking).

Another way to use the flowchart taxonomy is to see where there are similarities between pathways. For example, continuing with our biochar case, we can see that it is typically similar to forestry in its capture step but is very different from forestry in its storage step.

As we collect more information about the flow of carbon from left to right across this flowchart, we will be able to make the connecting lines scaled proportionally to the volume of carbon flowing through each connection. Sankey diagrams like this are already used to help people understand and make decisions about complex fields such as the course of primary energy supply to final energy demand. By building towards an analogous representation of the current and future CDR field, it will be easier to allocate limited resources (such as waste biomass) across categories and to understand the implications of supporting different processes across the three steps of the flow diagram.

The flowchart taxonomy is not meant to be definitive; instead, it is meant to show the power of looking at the same space with a new light. Other taxonomies and representations may be equally helpful for thinking about financing, policy, or RD&D in new ways. As the CDR ecosystem continues to mature and expand, researchers, funders, and entrepreneurs should consider the other ways in which the CDR ecosystem can be parsed, as these may lead to exciting new technical breakthroughs, commercial possibilities, and means of communicating specific messages to different audiences.