May 20 2022

Scope, goals, and implications of our recent seaweed farming modeling work

Freya Chay +Julianne DeAngelo 
Freya Chay

Julianne DeAngelo is a Ph.D. Candidate at the University of California, Irvine

Last month, we released an explainer article and interactive mapping tool on the potential of seaweed farming for both carbon removal and biomass products. The work was done as a collaboration between CarbonPlan and a larger academic research team led by Julianne DeAngelo, Steve Davis, and colleagues. The research team led the technoeconomic modeling and analysis, CarbonPlan built the interactive tool, and all of us collaborated to write about the work.

Since its release, there has been productive public conversation about the strengths and limitations of the modeling work. In particular, practitioners developing novel seaweed carbon removal approaches have pointed out that the cost ranges in the model may not reflect current advances taking place in the field, nor the potential for future innovation. Given our interest in both the research and practical implementation of carbon removal, we wanted to highlight this conversation and share some thoughts.

The technoeconomic modeling led by the academic research team relied on public, published data, which we think is critical for bringing rigor and transparency to these topics. Guided by the researchers’ ability to access public data, the structure of the model reflects some important assumptions about seaweed cultivation approaches. For example, the model assumes that seaweed is grown and harvested from an anchored floating array, which incurs both capital expenses (for buoys, anchors, boats, rope, etc.) and operating expenses (related to harvesting, transportation, and maintenance). While the model can accommodate a range of costs associated with each of these components, the structure of the model limits the set of practices to which it reasonably applies.

As a consequence, novel efforts which have not yet produced public data may fall outside of the model’s parameter space. The approach described by the company Running Tide, for example, avoids array-based infrastructure and does not require harvesting seaweed. Instead, they grow kelp on distributed floating buoys which are designed to subsequently sink without further intervention. If successful, such methods may well reduce costs outside of the modeled range per ton of seaweed grown and sunk. At the same time, it is unclear how free-floating arrays could be used to generate high-density yields, potentially necessitating a much larger area of overall cultivation to achieve climate targets. In the absence of public data on new cultivation approaches, it is not yet clear how to incorporate such methods into technoeconomic modeling of large-scale seaweed farming.

Despite the fact that our modeling effort did not directly describe novel approaches to seaweed cultivation, there are important takeaways that we think are transferable across approaches. Most prior technoeconomic analyses of seaweed farming for carbon removal have neglected important earth system considerations, such as the sustainability of available nutrients for gigaton-scale removal and the fraction of carbon taken up by seaweed that is actually removed from the atmosphere. The new model suggests these factors are critically important to understanding the potential for durable and verifiable ocean-based carbon removal, regardless of the farming method used.

In summary, we value how this modeling effort has relied on public information and stand by the key uncertainties and sensitivities highlighted. We also recognize that the reliance on public data implies that the modeling may not speak to some of the innovation happening right now. As this industry evolves, we hope that cost-effective and scientifically-rigorous solutions will continue to emerge in the space between what is currently published and what is possible.


We thank all authors of our original explainer article: Julianne DeAngelo (UCI), Steven Davis (UCI), Benjamin Saenz (, Isabella Arzeno-Soltero (UCI), Matthew Long (NCAR), Christina Frieder (SCCWRP), Kristen Davis (UCI), Kata Martin (CarbonPlan), Freya Chay (CarbonPlan), Jeremy Freeman (CarbonPlan), and Joseph Hamman (CarbonPlan).

Questions? Interested in collaborating on these problems?
email us
CarbonPlan is a registered nonprofit public benefit corporation in California with 501(c)(3) status.
SCROLL: 0.00