If you ask climate modelers how humanity can avoid severe global warming — say, 2°C or more — most will say we need to do two big things. First, we’ll need to reduce global greenhouse gas emissions down to zero by the end of the century. Second, since we’ve been so tardy in making those cuts, we’ll also need to figure out how to pull some carbon dioxide back out of the atmosphere.

And that’s … a problem. We at least have some notion of how to cut emissions. But sucking carbon dioxide back out of the atmosphere? At the massive scale likely needed? No one really has a clue how to do that. It’s a huge, embarrassing blind spot in climate policy.

The IPCC has estimated that, to stay below 2°C of warming, we’ll need to zero out our emissions and start removing between 2 and 10 gigatons of CO2 from the atmosphere each year by 2050. For perspective, all of the world’s forests and soils put together currently remove just 3.3 gigatons of CO2 each year. So imagine doubling or tripling that. Planting more trees could help, but we’ll need sweeping new carbon-removal techniques on top of that.

Right now, we have only crude ideas of what that might entail. Perhaps we could harvest trees sustainably, burn them for energy, and bury the resulting emissions underground — a technology known as bioenergy with carbon capture and sequestration (BECCS) that, in theory, is carbon-negative. Or we could try to boost the carbon-absorbing capacity of soil. Or we could deploy giant machines to suck CO2 out of the air (known as “direct air capture“). But we don’t yet know if these ideas are feasible. And surprisingly few people are even working on this.