“DAC removes 1.3 million tonnes per year. A single good year of global forest growth removes 10 gigatonnes. Why are we spending billions on machines when we could just plant trees?”
This framing shows up constantly, and it’s wrong in a specific way that’s worth unpacking — because the people making the argument often genuinely care about climate outcomes.
Afforestation and DAC don’t compete for the same slot in the climate portfolio. They address different risk profiles, operate on different timescales, and fail in different ways. The question isn’t which one to choose. The question is how to use each where it actually works.
Forests have constraints that tech CDR doesn’t. Land is the obvious one — meaningful afforestation at scale requires enormous areas, and those areas compete with agriculture, existing ecosystems, and human settlement. Permanence is another: a tree planted today is a carbon sink for decades, but it’s also a future wildfire, beetle outbreak, or drought waiting to happen. The 2023 Canadian fire season released over a billion tonnes of CO₂ from forest carbon that had been accumulating for years. Natural sinks aren’t guaranteed storage — they’re biological systems that respond to climate stress in ways we’re only starting to model well.
DAC, by contrast, can operate on degraded land with no soil requirements. The carbon it captures can be stored geologically in basalt formations where it mineralizes within years. That’s a different permanence guarantee than a living forest.
The timescale argument cuts both ways. Critics of DAC point out it’s still expensive and small-scale. True. But a forest planted today won’t hit significant carbon accumulation for 20-30 years. A DAC facility financed today can be operational within 2-3 years. Under scenarios where we need rapid carbon removal to hit specific temperature targets, that lead time matters.
The strongest argument for both working together: we’re in a position where we need every tool to work. We’re emitting ~37 gigatonnes per year. Natural sinks remove roughly 10. We’re not choosing between two options that could each solve the problem alone — we’re assembling a toolkit where every component needs to contribute.
Planting trees is good. Protecting existing forests is essential. Building DAC infrastructure is a hedge against the failure modes of both. These aren’t competitors. They’re the same team.