Carbon dioxide removal, often shortened to CDR, is any process that pulls CO2 out of the atmosphere and stores it somewhere durable, resulting in a net decrease in atmospheric carbon. It’s different from simply catching emissions before they leave a smokestack. CDR targets the CO2 that’s already spread into the air we breathe, and climate models suggest the world needs to remove roughly 6 to 10 billion tons of it per year by mid-century to stay on track for the goals set in the Paris Agreement.
How CDR Differs From Carbon Capture
The terms “carbon capture” and “carbon dioxide removal” get used interchangeably, but they describe different things. Carbon capture and storage (CCS) typically refers to grabbing CO2 at its source, like the exhaust stream of a cement plant or a natural gas power station, before it enters the atmosphere. It prevents new emissions but doesn’t reduce what’s already up there.
CDR is specifically about achieving a net reduction in atmospheric CO2. The carbon might be absorbed by trees, pulled from ambient air by machines, or locked into ocean chemistry. What makes it “removal” rather than just “capture” is the end result: the total amount of CO2 in the atmosphere goes down. Think of CCS as stopping the bathtub from filling faster, while CDR is draining water that’s already in the tub.
Nature-Based Approaches
The most familiar forms of CDR are biological. Planting trees, restoring forests, and managing soils to hold more carbon have been practiced for decades, and they still account for the overwhelming majority of CO2 removal happening today. Of the roughly 2 billion tons of CO2 removed annually worldwide, almost all of it comes from these conventional, land-based methods.
Afforestation and reforestation (planting new forests or regrowing old ones) can remove around 10 to 11 tons of CO2 per hectare per year, depending on the type of land and forest management practices. Soil carbon sequestration works differently: instead of growing wood, you change farming practices (cover cropping, reduced tillage, composting) so that more carbon stays locked in the ground. Blue carbon ecosystems, like mangroves, seagrasses, and salt marshes, store carbon in waterlogged sediments where it decomposes extremely slowly.
The tradeoff with all biological approaches is permanence. Carbon stored in trees can be released by wildfire, disease, or logging. Soil carbon can escape if land management changes. The most optimistic estimates put woody biomass storage at centuries and deep soil carbon at up to a thousand years, but those timelines depend on the ecosystem staying intact.
Direct Air Capture
Direct air capture (DAC) uses engineered systems to pull CO2 directly from ambient air. Large fans draw air across chemical sorbents, either liquid solutions or solid materials, that selectively bind with CO2 molecules. Once the sorbent is saturated, heat or pressure is applied to release the concentrated CO2, which can then be compressed and injected underground for permanent storage.
Liquid solvent systems typically use a potassium-based solution. Atmospheric CO2 dissolves into the liquid and forms a carbonate compound. Solid sorbent systems work by adsorption: CO2 molecules stick to the surface of a specially designed material and are later released when the material is heated. Both approaches produce a stream of nearly pure CO2 ready for storage.
The technology works, but the costs are steep. Current prices for direct air capture range from about $600 to $1,000 per ton of CO2 removed, making it far more expensive than nature-based methods. The installed capacity is also tiny. Novel CDR methods collectively, including DAC, remove only about 1.3 million tons per year, less than 0.1% of what conventional approaches achieve. Climate scenarios that limit warming to 1.5°C envision DAC scaling to anywhere from near zero to 1.7 billion tons per year by 2050, a gap that requires enormous investment in infrastructure and energy.
Bioenergy With Carbon Capture
Bioenergy with carbon capture and storage, or BECCS, combines two ideas. First, you grow plants or collect agricultural and forestry waste. Those plants absorbed CO2 while they were alive. Then you burn or process that biomass to generate electricity, heat, or biofuels. The CO2 released during combustion is captured before it reaches the atmosphere and injected into geological storage underground.
The result is energy production with a net-negative carbon balance. The plants pulled carbon from the air while growing, and that same carbon gets permanently buried instead of cycling back. Feedstocks range from dedicated energy crops to forestry residues and agricultural waste. Using residues is generally preferred because it avoids the need to clear land for new plantations, which would undercut the climate benefit.
BECCS is expected to play a major role in climate scenarios. Under pathways limiting warming to 1.5°C, IPCC models project BECCS removing a median of about 2.75 billion tons of CO2 per year by 2050, with some scenarios reaching over 9 billion tons. Its removal intensity per hectare tends to be higher than reforestation over long time horizons, though the exact numbers depend heavily on what feedstock is used and how efficiently the carbon is captured.
Ocean-Based Methods
The ocean already absorbs about a quarter of human CO2 emissions naturally. Ocean-based CDR aims to accelerate that process. The most discussed approach is ocean alkalinity enhancement: adding finely ground minerals (like olivine or limestone) to seawater to shift its chemistry. This increases the water’s capacity to absorb CO2 from the atmosphere and converts it into stable carbonate forms. A side benefit is that it reduces ocean acidification, which threatens coral reefs and shellfish.
Other ocean-based strategies include growing and sinking seaweed, and electrochemical processes that strip CO2 from seawater so it can absorb more from the air. These methods are still largely experimental, and questions remain about ecological side effects and how to verify that the carbon actually stays sequestered.
How Long Does the Carbon Stay Stored?
Permanence is one of the most important differences between CDR methods. Biological storage (forests, soils, wetlands) is inherently vulnerable. Trees burn. Soil gets disturbed. Even under the best conditions, carbon in woody biomass persists for centuries, and deep soil carbon can last up to a few thousand years, but those timelines assume the ecosystem remains undisturbed.
Geological storage offers much stronger guarantees. When CO2 is injected deep underground into rock formations capped by impermeable layers, it can remain trapped for thousands of years. Over time, the CO2 dissolves into surrounding fluids and eventually mineralizes, converting into solid carbonate rock. That mineral trapping can lock carbon away for tens of thousands of years. The most effective storage sites are those with thick, low-permeability seals or formations where mineralization happens relatively quickly.
This permanence gap matters for policy. A ton of CO2 stored in a forest that might burn in 50 years is not equivalent to a ton injected into basalt rock where it mineralizes over millennia. Some carbon credit systems are beginning to account for this by discounting shorter-duration storage or requiring buffer pools to cover potential reversals.
The Scale of the Challenge
The world currently removes about 2 billion tons of CO2 per year through CDR, almost entirely through tree planting and land management. Novel engineered methods contribute just 1.3 million tons, a rounding error against what’s needed. IPCC scenarios consistent with limiting warming to 1.5°C call for total CDR deployment in the range of roughly 6 to 18 billion tons per year by 2050, depending on how aggressively emissions are cut in other sectors.
Closing that gap requires scaling up every category of CDR simultaneously. Nature-based methods need to expand, but they face competition for land with agriculture and biodiversity conservation. DAC needs dramatic cost reductions and massive amounts of clean energy to power its systems. BECCS needs sustainable feedstock supply chains and CO2 storage infrastructure. Ocean-based methods need more research and regulatory frameworks that don’t yet exist.
CDR is not a replacement for cutting emissions. Every major climate assessment makes this point clearly: removal technologies supplement deep emissions reductions, they don’t substitute for them. The CO2 that doesn’t get emitted in the first place never needs to be removed. But for the emissions that are hardest to eliminate (from aviation, cement, agriculture) and for the excess CO2 already in the atmosphere, removal is increasingly viewed as necessary rather than optional.