Phytoplankton and carbon sequestration
Whatever you make doable ends up being done.
– Peter Fiekowsky
I recently read the book Climate Restoration, by Peter Fiekowsky, which is about rapid, aggressive carbon capture and sequestration. Fiekowsky suggests three techniques for CCS modeled on the natural biological and geological processes that Earth has used to regulate CO2 for hundreds of millions of years (and a bonus process for neutralizing methane). Fiekowsky believes we could reach 350ppm CO2 by 2050, and restore the planet to its pre-industrial sub-300ppm by the end of the century. And he’s probably right - with the caveat of unforeseen consequences.
The technique that really caught my eye, and the one he believes is most promising for maximum CCS at minimum cost, is triggering phytoplankton blooms in the open ocean by fertilizing the waters with iron. In theory, a mere ton of powdered iron (or iron oxide or iron sulphate), distributed over a wide area, could trigger a million tons of phytoplankton to bloom. This technique could capture tens of billions of tons of CO2 per year, on par with the 37Gt (billion tons) of fossil CO2 we currently release into the atmosphere every year. So you can see why this is a promising approach! It would be inexpensive, fairly low-tech, and work at incredible scales.
So, how feasible is this? And what are the implications? I started “doing my own research”. (Note: My analysis is heavily dependent on ChatGPT. I’ll include links to primary sources where I can, and I’m checking the reasoning within my own significant limitations, but your trust in my analysis should be bound by your trust in AI for this type of analysis.)
CO2 by the numbers
Before tackling phytoplankton, it’s important to understand the scope of the CO2 problem. Over the course of the past 10,000 years or so, the atmosphere has been about 280ppm CO2 - about two trillion tons of CO2. Earth had been under 300ppm for the past 800,000 years, sometimes dipping as low as 180ppm during glaciation periods.
Around 200 years ago, we started burning fossil fuel at industrial scale. As of December 2024, we are at 425ppm - the highest level in over three million years, a 50% increase in just 200 years. Another trillion tons. Half of this increase occurred in just the past 35 years (we passed 350ppm in 1988).
It seems likely to me that fossil fuel consumption is on the cusp of declining, and will approach zero by 2050, thanks to the ready availability of cheap solar power, batteries, and EVs. But Net Zero is only one step on the path to actually undoing the damage. If we don’t clean up the CO2 we’ve already spewed, climate change will continue to wreak havoc for tens of thousands or even millions of years. Leaving the atmosphere at near 500ppm is no decent way to declare victory.
To restore a pre-industrial atmosphere, we need to remove a trillion tons of CO2, an amount that took 200 years of industrialization to produce. Not an easy task.
Phytoplankton by the numbers
I proved that to be a scientific impossibility seven times!
– Plankton, Spongebob Squarepants
Phytoplankton accounts for about half the biomass that grows on Earth each year. That’s not because there’s a lot of it! There is only about 300Mt of it at any given time, little more than half a percent of Earth’s 500Gt of living biomass. But phytoplankton grows very fast, and only lives for a few days. So Earth produces about 50Gt of phytoplankton every year. As phytoplankton dies, much of it sinks to the bottom of the ocean, where its carbon is sequestered indefinitely, anywhere from thousands to millions of years. Petroleum is the result of millions of years of accumulated phytoplankton, buried under sea floor sediments.
That 50Gt of phytoplankton, in turn, consumes over 190Gt of CO2 via photosynthesis. If this seems odd to you, remember that 73% of the mass of CO2 is oxygen, which is released by the photosynthesis process. Most of that is simply returned to the environment by the chain of life, as other marine organisms consume phytoplankton and exhale CO2, or pass the carbon on to organisms that consume them.
Phytoplankton permanently sequester between 19.25 and 38.5 gigatons (Gt) of CO₂ per year in the deep ocean (according to ChatGPT’s calculations). This represents 10-20% of the total CO₂ they capture annually through photosynthesis. But keep in mind that this level of sequestration is the normal level of the past couple million years. This isn’t offsetting the release of fossil carbon by human activity at all!
And finally, keep in mind that phytoplankton are the bedrock of the entire oceanic food chain. Phytoplankton need nothing more than seawater (and its dissolved CO2), sunlight, and a few trace micronutrients - like iron. This is where iron fertilization comes into play. Phytoplankton are consumed by zooplankton, which are consumed by fish and whales, which are in turn consumed by other marine animals, and so on, all the way up to alpha predators like sharks and humans.
How iron fertilization works
“Give me half a tanker of iron, and I’ll give you the next ice age.”
– John Martin
Iron is present in seawater, but it is not anything close to evenly distributed. There is over 100,000 times more iron dissolved in coastal waters than in some parts of the open ocean. That iron comes from dust carried on the wind from deserts, from volcanic ash, or gets stirred up from the muck on the ocean floor.
Large parts of the open ocean are basically a desert - not because they’re dry, but because they’re lifeless. There is virtually no phytoplankton in these ocean deserts, so there is not much other life, either - phytoplankton is the base of the food chain. There’s nothing to eat in these ocean deserts. But there is seawater. There is sunlight. Why isn’t there phytoplankton? Because the waters are missing a key micronutrient - iron.
This theory was originally developed in the 1930s by biologist Roger Hart, but it was not tested. In the early 1970s, the theory was taken up in earnest by John Martin, the head of Moss Landing Research Laboratories in Monterey, CA. He developed new techniques for measuring iron concentrations in seawater. He tested the hypothesis, first by adding iron to seawater samples and comparing their phytoplankton growth relative to untreated samples, then testing the theory in the open ocean - an experiment he did not live to see himself, but which was completed by his colleagues shortly after his death in 1993.
Dr. Martin’s experiments, and followup experiments conducted since then, have proven the simplicity and effectiveness of iron fertilization for creating massive plankton blooms. But so far, there has been almost no motion toward implementing iron fertilization at scale. Instead, billions of dollars have been squandered on ineffective direct air capture machinery that has done little but generate CO2 to be injected into oil wells to improve their yield - the opposite of what needs done! Billions more go to greenwashed “carbon offsets” that cannot even begin to offset the carbon we continue to produce. Meanwhile, well-meaning environmentalists oppose any sort of carbon sequestration, for fear that it will be used as an excuse to continue burning fossil fuels.
The first catch: the limits of usable ocean eddies
Eddy? That’s a rather tender subject.
– Frankenfurter, The Rocky Horror Picture Show
So why not just dive in, as Fiekowsky suggests? Because we would be messing with one of the most fundamental systems of life on Earth. And it’s quite possible that iron fertilization would not be as effective as we hope.
We need to keep the whole system in mind. The goal is not to stimulate phytoplankton growth - the goal is to sequester carbon, by sinking it to the bottom of the ocean, effectively removing it from the active biosphere. Not all phytoplankton sinks to the bottom. The majority gets eaten by zooplankton and other marine life, keeping it in the biosphere. In worst-case scenarios, the bloom may end up putting more carbon in the atmosphere, not less.
The ocean is full of eddies, “ocean pastures” that support life. Eddies can be anywhere from a few kilometers to over 100 kilometers across. But not all eddies are the same! First and foremost, there are cyclonic eddies, and anticyclonic eddies, split approximately 50/50. Cyclonic eddies have a cold core. They lift colder, nutrient-rich water from deeper waters to the surface. These tend to have natural phytoplankton blooms and can be teeming with life.
Anticyclonic eddies, on the other hand, trap warm, nutrient-poor water at the surface. It would seem that anticyclonic eddies would be better candidates for iron fertilization, but that’s not necessarily the case. The downwelling of such eddies would not drive organic material deep enough, so it stays near the surface, where it eventually returns to the biosphere. And mid-depth decay can create oxygen-starved “dead zones”, which are harmful for ocean life in general.
In addition, lifespans of eddies vary wildly, from hours or days to months, or even over a year. Short-lived eddies are not worth the effort of seeding.
Luckily, there are thousands of eddies in the ocean at any given time. To do iron fertilization effectively, we need eddies that are:
- large
- cyclonic
- long-lived
- rich in phosphorus and nitrates, but iron deficient
Such HNLC (high nutrient, low chlorophyll) regions are plentiful, but hardly the entire ocean. This probably puts a cap on the effectiveness of iron fertilization. Let’s try to estimate that. A powerful phytoplankton bloom, caused by iron fertilization in a HNLC zone, might contain 50-200 tons of plankton per km^2. This represents 185-735 tons of CO2. But most of it will not be sequestered, but rather will return to the biosphere. So in practice, a 1000km^2 iron-induced bloom could permanently sequester up to 200,000 tons of CO2. For this to scale to anything resembling the kinds of goals Fiekowsky is talking about, we’d need to trigger 100,000 such blooms a year, in order to sequester 20Gt of CO2. That’s about half a year of our current fossil CO2 output. Is that even possible?
Let’s say we could get 10 blooms out of a given eddy. This would require 10,000 eddies a year. And I think 10 blooms is the most we could expect any single ship to produce in a year, considering the monitoring and data collection, plus travel and dock times. So we’d also need 10,000 ships and crews. Not impossible, but not something that could easily be done, either. That’s about half the high-water fishing boats on Earth today.
Worse, there are only hundreds of large scale cyclonic eddies globally in HNLC areas at any given time. That’s not nearly enough to support the scale we are talking about! Of course, we can expand our scope, triggering blooms in areas with less potential, but the problems remain. Capturing 20Gt, or even 10Gt of CO2 annually using iron fertilization is, at best, a stretch goal. Under 5Gt/year is probably more realistic.
The second catch: impact on marine life
Increasing phytoplankton enough to sequester tens of billions of tons of CO2 per year would mean a 50-100% (or more) increase in total phytoplankton. Since phytoplankton is the foundation of the oceanic food chain, this would have tremendous impact on the entire ocean biosphere, favoring some species and challenging others. This work would almost certainly need to be done exclusively in blue-ocean waters, away from the rich coastal shelves we depend on for most of our human relationships with the ocean. And it would need to be done for decades, even centuries. This would be met with resounding hostility from most environmentalists and probably most scientists.
This much additional plankton would lead to more oceanic oxygen depletion, harmful algae blooms, jellyfish, and other issues. It may also lead to nutrient depletion of essential phosphorus and nitrates. I do not feel even remotely qualified to speak on this, but it’s very concerning.
What are the alternatives?
The first and most obvious is to do nothing. But as I said earlier, doing nothing to sequester CO2 is morally reprehensible. On the other hand, doing nothing is probably the most likely avenue, considering the political, economic, and long-term sustainability challenges this effort would face.
The second is to scale up slowly, proving the effectiveness of the approach and carefully monitoring its impacts on oceanic life. Most likely, it will work well up to a certain point, then start causing issues too big to ignore. That point is the point of sustainability. It might be 10Gt, or less than 1Gt. There really no way to know. Such scaling up would likely take decades. Limits would also be regional, with some regions more suitable (or less sensitive) than others.
And above all, we need to reduce the amount of CO2 we are producing, as quickly as possible! The more CO2 we add, the less effective iron fertilization (or any other sequestration) will be. They would just be offsetting current emissions, not reducing overall CO2 concentrations. If we don’t stop burning fossil fuels, we’re screwed, regardless of CCS.
Conclusion
Iron fertilization could still be one of our best approaches for CCS. It’s inexpensive and low-tech (even at maximum scale, it would probably cost only tens of billions per year). At the right scale, it could have positive effects on ocean life.
But it’s not a silver bullet. It’s not going to easily reduce CO2 to pre-industrial levels by the end of the century. More likely, we need to do every viable approach, and even then, we may be looking at centuries of effort to decarbonize the atmosphere.
But, as Wendell Berry wrote, “This would be work worthy of the name ‘human’.”