The Measurement Problem at the Heart of Carbon Removal

In September 2025, NASA's Perseverance rover photographed small red spots ringed with dark material on a Martian rock — "leopard spots" that, on Earth, form either through hot acidic chemistry or through biological action. Scientists couldn't rule out life. The reason this made headlines wasn't just the tantalizing implication. It was that Perseverance couldn't identify the organic molecules it found. The detection worked. The interpretation remained genuinely open.

That's a measurement problem. And it turns out the most consequential measurement problem happening right now isn't on Mars — it's in our oceans.

What You Can't Count, You Can't Sell

Ocean-based carbon dioxide removal is one of the more promising levers we have for pulling CO₂ back out of the atmosphere. The ocean already absorbs roughly a quarter of human emissions annually. The question is whether we can enhance that capacity deliberately — through iron fertilization, kelp farming, alkalinity enhancement, or other approaches — and do it in a way we can actually verify.

Here's the problem: the ocean is not a controlled experiment. Carbon moves through it in three dimensions, across temperature gradients, through biological systems that are eating and dying and sinking at rates we can only partially track. When you add something to the ocean to sequester carbon, determining how much carbon actually stayed sequestered — versus getting outgassed somewhere else, or taken up by a different part of the food web — is genuinely hard. Not hard in the "we need better computers" sense. Hard in the "we don't fully understand the system" sense.

This is what the CWorthy initiative is trying to crack. It's a convergent research project focused specifically on building the measurement, reporting, and verification frameworks for ocean CDR — because without those, you can't have carbon markets, you can't have accountability, and you can't scale anything responsibly. The science has to come before the commerce, or you end up selling air.

The Infrastructure Nobody Wants to Fund

This is where behavioral economics enters from an unexpected angle. One of the consistent findings in that field is that humans are bad at valuing things with diffuse, delayed, or uncertain payoffs — even when the expected value is clearly positive. We discount the future steeply. We demand legibility. If we can't count it cleanly, we mentally round it toward zero.

Carbon measurement infrastructure is exactly the kind of thing that suffers from this bias. It's not a product. It doesn't remove any carbon itself. It produces data — and data about a process happening in an ecosystem that most people will never see. Funding it feels abstract. The Carbon Containment Lab's approach of building physical mineralization test beds across the country is partly a solution to this: make the experiments real, make them legible, make them something a startup or a regulator can point to. Tangibility is a nudge.

But the deeper issue is that the entire carbon removal sector is trying to build markets for something humans are cognitively poorly equipped to value. Carbon in the atmosphere is invisible. Its effects are distributed across decades and geographies. The damage it causes is statistical rather than personal. Every feature of that description is, from a behavioral economics perspective, a recipe for systematic undervaluation.

Nudge theory suggests you can redesign choice architectures to compensate for these biases. Default enrollment in green energy programs works. Showing people their energy use relative to neighbors works. But carbon markets aren't consumer products — they're institutional mechanisms, and the biases they need to overcome are operating at the level of investment committees, government procurement, and corporate sustainability departments. The nudges required are structural: mandatory accounting standards, third-party verification requirements, liability for false claims. Less "gentle push," more "changed rules of the game."

Verification as the Actual Hard Problem

What connects the Mars biology question to the ocean carbon question is that both are fundamentally about what counts as evidence. On Mars, we can detect organic molecules but can't identify them. In the ocean, we can measure carbon fluxes but can't always attribute them. In both cases, the limiting factor isn't ambition or even technology — it's our ability to interpret what we're observing with enough confidence to act on it.

For climate, the stakes of getting this wrong are asymmetric. If we overcount carbon removal and build policy around inflated numbers, we've bought ourselves a false sense of progress while the atmosphere keeps warming. The scientific community working on ocean CDR knows this, which is why the measurement-first approach isn't excessive caution — it's the only way to avoid building an elaborate structure on a foundation that hasn't been checked.

The unglamorous truth about scaling climate solutions is that a lot of the real work right now looks like infrastructure: test beds, verification protocols, measurement standards, data pipelines. None of it removes a single ton of carbon on its own. But without it, you can't know whether anything else does either.

Perseverance is still drilling. The data is still coming in. Certainty, when it arrives, will have been earned slowly — and that's not a failure of the mission. It's what honest science actually looks like.