Can ocean-based carbon removal deliver measurable climate impact at scale?

How much carbon can ocean farming actually pull from the air?

Macroalgae (seaweed) farming can measurably improve local seawater chemistry. In a 2025 field study in Yueqing Bay, China, farming the seaweed Porphyra haitanensis raised pH by 0.09 units and lowered the partial pressure of CO₂ (pCO₂) by 118 micro-atmospheres across kilometer-scale areas—a clear sign that the seaweed was pulling carbon out of the water [1]. A different species, Hizikia fusiformis, produced more localized but still significant effects, cutting dissolved inorganic carbon by 1.84 mg per liter and pCO₂ by 82.6 micro-atmospheres within 100 meters [1].

But local water chemistry changes don't automatically translate into permanent atmospheric CO₂ removal. A 2023 global modeling study found that if macroalgae were grown at a massive scale—0.5 billion tons of carbon per year—the actual enhancement of the ocean's ability to absorb CO₂ from the air would be only 0.39 billion tons of carbon per year, an efficiency of 79% [2]. When the model also accounted for nutrient limitations (seaweed needs nitrogen and phosphorus to grow), the realized carbon removal dropped further: only 0.21 billion tons of carbon per year, or 58% of what the seaweed initially took up [2]. The rest of the carbon simply wasn't replaced by new CO₂ from the atmosphere.

What's the hidden cost? Ocean oxygen loss and food web disruption

The biggest catch with biotic ocean carbon removal—methods that grow and sink organic matter—is that they can worsen ocean deoxygenation, which is already a major problem from global warming. A 2025 review of multiple marine carbon dioxide removal (mCDR) options found that biotic approaches like macroalgae cultivation and sinking can cause a loss of dissolved oxygen that is 4 to 40 times larger than the oxygen gain that would result from the cooling effect of the CO₂ removed [4]. In other words, the cure could be worse than the disease for ocean oxygen levels.

This oxygen loss happens because when the seaweed or other organic matter sinks and decomposes, microbes consume oxygen to break it down. The same review notes that geochemical approaches (like adding alkaline minerals to the ocean) and biotic methods that avoid decomposition in the water column can have minimal oxygen impacts [4].

There's also a direct hit to marine food webs. The 2023 modeling study found that large-scale macroalgae farming could reduce phytoplankton primary production—the base of the ocean food web—by up to 40 grams of carbon per square meter per year in the eastern tropical Pacific [2]. That's a serious concern for fisheries and marine biodiversity.

Which methods and locations work best—and which don't?

Not all ocean carbon removal methods are equal. The 2025 review on oxygen impacts draws a clear line: biotic approaches that involve sinking and decomposing biomass are the riskiest for oxygen, while geochemical methods like ocean alkalinity enhancement (adding crushed minerals to seawater) can be applied with minimal oxygen side effects [4]. A 2026 modeling study of ocean alkalinity enhancement found that it could raise the ocean's aragonite saturation state (a measure of ocean acidification) from dangerous levels back toward safe planetary boundaries—from about 2.6 to 2.97 by 2300, close to the 2.75 boundary [3]. That same study showed that reforestation on land had a similar effect on CO₂ concentrations, but ocean alkalinity enhancement had negligible impact on land use [3].

Location matters enormously. The 2023 modeling study identified eastern boundary upwelling systems, the Northeast Pacific, and the Southern Ocean as potentially promising regions for macroalgae-based carbon removal, because these areas have the nutrient supply and ocean physics to support efficient CO₂ uptake [2]. But the same study warned that in other regions, the nutrient competition between seaweed and phytoplankton could actually cause a net reduction in the ocean carbon sink [2].

About These Sources

This answer is built on 5 peer-reviewed studies — published from 2023 to 2026, 4 from 2024 or later, 3 in Q1 journals — selected as the most relevant from 5 studies that passed quality screening, drawn from 56 papers retrieved from a database of over 500 million.

Sources used in this answer

Related research questions