Genetically engineered phytoplankton might be used to photosynthetically convert CO2 into an ultra-stable carbon sequestration medium. A biological approach allows for a programmable and self-replicating technology that can work at any scale. When coupled with photosynthetic carbon fixation, genetically engineered organisms present an incredibly scalable approach for Carbon Dioxide Removal (CDR). We propose to take advantage of the vast, unused photosynthetic potential of our oceans with genetically engineered carbon sequestration phytoplankton.

The problem is that most of the good places to do photosynthesis are already taken. This map shows where photosynthesis is most and least active. As shown on the map, large portions of the ocean lack the nutrients necessary for photosynthesis. These same areas receive a lot solar energy and could be an incredibly powerful approach for CDR. New biotechnology approaches could dramatically increase the photosynthetic capacity of our planet. This section explores how we might use existing biology in new ways or even create new biology to turn the ocean into a giant, light-driven carbon sink using genetically engineered phytoplankton.

How much CO2 it could absorb if everything worked perfectly?

If we wanted to remove about 47 gt/year, we would need about 1% of the ocean's surface assimilating CO2 at full speed.
47Gt CO2
per year
Here's the maths

The fastest recorded CO2 fixation by existing/known biology is from photosynthetic algal beds, which can fix up 4.4 kg of carbon or about 16.3 kg of CO2/year. We will be conservative and say that our algal beds fix 2.5 kg of C per square meter per year and 2.5 kg of C works out to 9.167 kg of CO2. So, if we can fix 9.167 kg of CO2 per square meter per year, how big an area would we need?

The ocean's surface area is 5.1E8 square kilometers * 1E6 square meters/sq km = 5.1E14 square meters.

If we multiply this times the maximum CO2 assimilation rate of algal beds which we will say is 9.167 kg of C per square meter per year, we could assimilate about 4.67E15 kg CO2 per year

This works out to ~4675 Gt of CO2/year if the entire ocean is sucking up CO2 at full speed (this estimate ignores quite a few details)

Luckily we only need to sequester 20 to 40 Gt.

* 4.67E15 kg = 4.67E12 tonnes = 4.67E9 kilotons = 4.67E6 megatons = 4.67E3 tonnes

Why isn't there already life in the open ocean?

If there is all this free space to grow, why aren't the oceans full of photosynthetic life already? Turns out that the surface of the ocean is low in a few trace minerals that are critical for microbial growth and photosynthetic activity. Apart from those minerals, the ocean has everything needed to solve this problem: a lot of water, sunlight, and CO2. If we could figure out a way to solve phytoplankton's mineral requirements, the entire ocean could become a powerful carbon sink, as well as a new frontier for economic growth.
There are two ways of solving this problem - either get the missing minerals where they are needed (fertilization) or eliminate the need for minerals all together. We will start with what's possible today and wade into what some might consider sci-fi. Either way, current biology won't do the trick.

In-depth description & science

Grow and dump/boat-based cultivation and distribution

The most straight forward approach for using a phytoplankton strategy is to treat it as a massive logistics problem: how do we make and distribute enough phytoplankton to put a dent in CO2. For example, we could imagine using 1000 ships towing giant photosynthetic bioreactors that are constantly generating a genetically engineered phytoplankton. Each time the phytoplankton population doubles (which can be as fast as every 2 hours in optimal conditions) [1], half could be dumped overboard where the microbes would switch out of "growth mode" and into "CDR mode."

Phytoplankton have been engineered to produce a variety of different materials renewably. We imagine a phytoplankton engineered to convert sunlight and CO2 into a chemically stable bioplastic. Once the phytoplankton is in carbon sequestration mode, they would put all of their photosynthetic energy into sequestering CO2 into a stable bioplastic/carbon sequestration medium. This material could sink to the bottom of the seafloor where it would remain sequestered or be harvested.

While most other ideas involve a lot of scientific research, this idea is foremost an engineering/infrastructure problem (though engineering a phytoplankton capable of driving CO2 into bioplastics will require development). There are a few key questions to consider.

First, the ocean is huge. In order to cover the surface with enough phytoplankton, we would need a lot of boats. But how many boats? Second, it is important the the phytoplankton be spread out. If they are too close together, then we will only get a local depletion of CO2. Third, how long would the phytoplankton be actively able to sequester CO2 before dying? Fourth, what happens to all this biomass?

These are tough questions to answer and these are just the first order considerations. But to get the ball rolling we have put together a framework that could help drive thinking forward.

[1] Also, phytoplankton grow super fast with extra CO2. So waste CO2 from powerplants could be bottled up and used to accelerate the growth of these bugs.

Complex behavior/self-assembling and self-replicating, self-fertilizing microbial mega structures

Bacteria can be programmed with complex behavior [2]. It might be possible to apply more complex approaches to phytoplankton. Imagine running a "biological python script" on phytoplankton. This could open up all sorts of interesting applications. For example a phytoplankton might be programmed in such a way that they could migrate to collect necessary nutrients or self-assemble into gigantic oceanic infrastructure using nothing but CO2 and sunlight.

So, let's imagine a phytoplankton that is able to grow in a bunch of different "modes." They can grow as free-swimming bugs, or they can aggregate into a big, floating, photosynthetic lily pad like entity. In lily-pad mode they would make a great self-replicating CDR system until they ran out of mineral nutrients. There are a ton of nutrients on the ocean floor. Our lily pad might decide to change buoyancy and sink to the bottom of the ocean to get more nutrients.

Another possible approach that could make use of the same basic concept of complex cellular behavior is to create phytoplankton that self-assemble into gigantic, megastructures. This could probably even be done off shore where the biology would be easy to fertilize. A few kilotons of phytoplankton could grow and swim around until they decided to switch into self-assembly mode. They might self-assemble into a massive kilometers long tube. Imagine a tube of 1000 m tall, with walls that are 0.1 meter thick and about 1 meter diameter (pi ~3 meters in circumference). This would work out to about 300 cubic meters worth of material or about 0.3 kilotons.

This technology could be programmed to have some really useful behavior. One side of the massive tube could float towards the sea surface and sprout leaf-like structures, while the other side might sink to the bottom of the ocean and attach itself into the ocean floor, putting out "roots." We would have essentially built a primitive tree-like entity.

If we build it right, the hollow center, might be able to draw up mineral nutrients from the ocean floor and deliver them to the the photosynthetic surface of the sea. This could fertilize the growth of more self-assembling mega-structures. Depending on nutrient availability, these entities could be engineered to create all sorts of interesting structures in the open ocean and could even create whole new ecosystems for marine wildlife.

[2] When people talk about programming biology, they usually mean re-writing DNA to add some new feature. In this context, when we are talking about programming biology, we mean actually writing a complex, responsive biological programs with if/else statements, inputs and outputs that respond to a variety of different signals etc. We imagine something like a biological or biochemical Turing machine/computer instantiated at the level of a microbial cell. Actual programming. We imagine something like a more engineering or applications-oriented version of systems biology, where something approximating a multi-line python script could be instantiated in an organism.

Mineral-free biology

Up until now, we have stayed within the limits of known biology. Everything up till now has precedent. But we want to think big. So we are leaving precedent behind and trekking into future technologies.

As we mentioned, mineral limitations are a major problem for using phytoplankton as a CDR tool. Many of these minerals are metals. These metals are used in various enzymes called metalloenzymes. All organisms have metalloenzymes. This may be leftover from life's early evolution and may not necessarily be a necessity for biochemistry. For examples, the organisms Borrelia burgdorferi (lyme disease) and Lactobacillus plantarum (yogurt) have eliminated all iron from their metalloenzymes and instead use a different metal called manganese. This suggests that there is some flexibility with how metalloenzymes do their job. It might be possible to eliminate metals from these metalloenzymes all together. If we could eliminate biology's metal requirements, phytoplankton might be able to grow in photosynthetic dead zones that have everything needed for CDR except for metals.

However, even if we were to remove all metals from biology, other nutrients would eventually become rate limiting. It is not possible to eliminate phosphorous from life because phosphorous is central to so many parts of biochemistry: chemical backbone of DNA and RNA (called phosphodiester bonds), ATP, the phospholipid bilayer, and many more.

The question then becomes, whether we can create new genetic chemistries that are not based on DNA, but some new genetic polymer? New proteins? New metabolism? Can we assemble these components into a form of biology that only needs CO2, nitrogen, water, and sunlight? Solving this problem is probably not feasible without major advances in biomolecular engineering. However, if we were able to create such a system, it would present a very effective tool for CDR.

What makes it a frontier technology?

The strategies range from requiring extensive protein, metabolic, and genetic engineering of phytoplankton to building an entirely new life form from scratch. While the metabolic engineering described for phytoplankton that convert CO2 into a storage medium has a lot of precedent, new systems of distribution and cultivation will be required. While simple genetic circuits and logic have been engineered into a variety of cells, much more complexity is required before we can engineer the sort of complex behavior required for self-assembling megastructures. The idea of a second genesis, building an entirely new form of biology and biochemistry from scratch most certainly seems like science fiction, but labs are already starting to scratch the surface (reference section) and we are looking forward to what these folks will make. Finally, the use of genetically engineered bacteria for biological production directly in the environment is not a well studied discipline.

Challenges - what is the non-starter:

All of these approaches come with significant financial and technical risks. Even if all the technology was developed, it is entirely possible that the solution might not perform as expected. More modeling is required to get a higher resolution answer on this. In addition, all of these approaches rely on the release of genetically engineered phytoplankton into the ocean. Some folks might not be too keen on that idea.

For building a giant fleet of phytoplankton boats, there are no total deal-killers apart from the scale and cost of the approach. However, there are 12.5 million boats registered in the USA, so a fleet of 1000 boats might not be too unreasonable. On top of this, we would need to build huge bioreactors that are transparent, can survive rough seas, and generate many tonnes of phytoplankton. Finally, while engineering a CDR biosynthetic pathway into phytoplankton is possible, it is not clear how performance would be impacted upon release into the ocean.

For engineering complex behavior into phytonplankton, while it would not be easy, there are no impossible technical hurdles. However, the fact remains that nothing like this has yet been accomplished [The road to synthetic multicellularity 2018]. Multicellularity is complex, but has come about many times throughout life's history on Earth. A key component of self-organization is cell to cell communication systems; while some such systems exist, news systems would be required. In addition, an extracellular matrix would likely be required. This allows cells to stick to one another and can be thought of as the scaffolding of multicellular organisms.

Re-building biology from the ground up is likely impossible at the moment. For all we know, various aspects of biology could be critical in a way that we do not necessarily understand, constraining the engineering options. Another thing to keep in mind is that life evolved only once (as far as we can tell). Oxygenic photosynthesis evolved only once in Earth's history. Therefore, attempting to re-build two of the most unlikely events that ever occurred in the history of life will probably be pretty tough. Many of the key features of biology were one time events. We probably don't have the ability to precipitate a second genesis today, but if someone has some interesting results, get in touch!

What research needs to be done and which type of researchers would be required

All of these approaches require the release of heavily genetically engineered organisms into the environment which comes with a host of its own issues. Engineering biological control systems that function at oceanic scale is critical for the use of the broad environmental release of synthetic biology.

Building complex behavior into biological systems requires complex control elements. In computers we use little switches called transistors. Engineering complex behavior into phytoplankton will require many biological switches that can be combined to create artificial biological behaviors.

Creating new biochemistries from the ground up requires enormous leaps in our ability to build biomolecules. Engineering such systems is clearly beyond today's technology, but perhaps new computational approaches that can gain and exploit a general intuition of the interplay between biomolecules and biochemistry might allow us to develop biological systems that are otherwise unreachable by abiogenesis/natural evolution.

Energy requirements

Phytoplankton CDR would be solar powered, so the energy costs of running the phytoplankton would be free. For the grow and dump approach, the boats needed to distribute the phytoplankton could be solar or nuclear powered.

For the other approaches where minerals are not required, then the system is fully self-sufficient and no additional energy is required.

Connection to Other Ideas

A genetically engineered organism deployed at large scale could provide a programmable, biological interface to modulate the environment. By building control switches into these organisms, we could essentially create a user interface for the environment. We might be able to control weather patterns by adjusting phytoplankton behavior. This could be used to increase rainfall in various areas and possibly help desert flooding.

Possible Business Models

Right now there is not too much economic activity on the ocean other than fishing, shipping, and oil drilling. This all stems from the fact that phytoplankton are the foundation of the food chain. By figuring out how to grow phytoplankton on the open ocean, we could vastly increase our planet's biological productivity. This could generate a useful biomass byproduct that might help to power growing economies.

Grow and dump could be used in conjunction with carbon credits. Power plants could both capture their CO2 and then use that CO2 to build phytoplankton biomass, which could then pull even more CO2 out of the atmosphere.

Estimated Costs

Phytoplankton is a very attractive CDR strategy because it is a set it and forget it approach. They can be thought of as programmable, self-replicating, photosynthetic, CDR factories.

The biggest strength is that this approach exploit one of biology's many super-powers: self replication. Once it is built, it can generate additional copies of itself. This means there is no infrastructure requirement and no energy requirements.

The real cost would be the development/bioengineering of such biotechnologies, an undertaking that could run into the billions depending on the approach. However, there are many synthetic biology startups that have successfully built and scaled novel biosynthetic pathways for a few million.

For the grow and dump approach, the cost depends on how much phytoplankton we need to grow. If our phytoplankton is able to sequester 100X, then we can probably get away with 1000 boats towing massive kilometer-long bioreactors. If each of these boats costed about $10 M, we might be able to build a fleet for ~$10 B.

Building self-assembling megastructures is probably a few orders of magnitude more difficult. New switches, sensors, and control systems, cell-cell communication, extracellular matrices, and many more components would need to be created. If we say there are 10 systems that need to be built from scratch at a cost of $100 M each, that would put us at $1 B in parts. Assembling those parts would probably take another $1 B. Optimizing the system would probably take another $1 B.

Re-engineering life from scratch would be the biological equivalent of going to the moon. We probably won't get there within the next five decades, which makes it really tough to estimate the price. However, the incredible trove of insight, techniques, and new technologies would super-charge the biotechnology industry allowing humanity to build even more ambitious, large-scale biological systems that might even be used to terraform planets one day.


The phytoplankton CDR approach raises some big risks as it is essentially a geoengineering project. On top of this, the broad environmental release of genetically modified organisms raises concern from some folks. Figuring out how to control these massive biological systems will be critical for their safe and effective use. Another big risk is what to do with all the captured carbon. Where does it go? Is it actually harmless if it sinks? Will it wash up on beaches?

While there are many questions that must be asked, answers with any degree of certainty will probably be tough to find. Perhaps the most important questions is whether the existential threat of climate change is worse than unknown potential consequences of CDR. Our planet is at a tipping point. The window for easy solutions is already closed. Doing nothing is guaranteed suicide. Therefore, we have two options: we try leverage new technologies for CDR (though there will probably be environmental and human casualties), or we sit on our hands and watch the planet die.

What is the 2040 vision for the idea? How would it work 20 years from now?

Earth is full of uninhabited niches that current biology cannot get a foothold due to a few missing nutrients. By engineering new biological systems that are capable of surviving in otherwise desolate areas, we might create new sources of productivity. We might use CDR to make the world a more bountiful place and shrink the wealth disparity resulting from material scarcity.

Additionally, such projects could lay the biotechnology groundwork for engineering organisms capable of surviving or even flourishing in environments outside of earth. In building the technologies to save our planet, we might create organisms to terraform Mars, the moons of Saturn or Jupiter, or maybe even organisms that could colonize asteroids. But to reach the stars, we need to save our planet.

2018 2040


Stal L. J. (2012) Cyanobacterial Mats and Stromatolites. In: Whitton B. (eds) Ecology of Cyanobacteria II. Springer, Dordrecht

John Bauld BMR Journal of Australian Geology & Geophysics, 6, 1981, 307-317 Geobiological role of cyanobacterial mats in sedimentary environments: production and preservation of organic matter