The Ocean’s Carbon Vaults Are Far Bigger—and Far Weirder—Than Climate Accounting Admits

For two decades, climate negotiators and carbon markets have treated mangroves, salt marshes and seagrass meadows as the ocean’s flagship carbon vaults. A sweeping new synthesis argues that the real blue carbon ledger is far larger, far stranger and far more imperiled than those three names suggest. Writing in the open-access journal Environmental Advances, researchers have distilled 3,033 peer-reviewed publications into one of the most comprehensive portraits yet of how the sea captures, transforms and entombs carbon—and their verdict is blunt: kelp forests, oyster reefs, coral reefs and the ocean’s invisible microbial machinery all move the planet’s carbon budget in ways that official accounting has largely ignored, and overlooking them, the authors contend, is a disservice to climate science itself.

The stakes could hardly be higher. The oceans absorb roughly 30 percent of the carbon dioxide humanity emits, and more than half of the planet’s photosynthetic carbon capture takes place in seawater rather than on land. Because water holds heat with extraordinary efficiency, the ocean has also soaked up the bulk of the excess energy trapped by greenhouse gases, and the resulting warming and acidification now threaten the very carbon-handling machinery the review describes. Vegetated coastal habitats occupy only about 8 percent of the ocean’s surface, yet they account for almost half of all carbon buried in marine ecosystems, with per-area burial rates approaching 200 times those of the open ocean. Coastal zones generate roughly 20 percent of the ocean’s organic matter and receive a riverine subsidy of about 426 teragrams of carbon each year—around 60 percent as dissolved organic carbon and 40 percent as particulate organic carbon—blurring the line between “green” carbon washed off continents and “blue” carbon fixed at sea.

The paper itself is a feat of bibliometric cartography. Following the Preferred Reporting Items for Systematic Reviews and Meta-Analysis protocol, the team searched PubMed, Web of Science and ProQuest with combinations of terms such as “blue carbon,” “marine carbon sequestration” and “vegetated coastal.” The initial trawl returned 7,589 records; after duplicates were removed, titles, abstracts and full texts were screened for relevance, and 28 papers were added manually, 3,033 publications spanning January 2009 to July 2024 formed the final corpus—the field effectively begins in 2009, when a United Nations report formally coined “blue carbon.” Mapping the corpus with VOSviewer revealed six keyword clusters orbiting blue carbon, climate change, carbon sequestration, the carbon cycle, ecosystems and economic valuation. A timeline analysis showed terms such as “carbon dioxide removal” and “carbon stocks” surging only in recent years, a signal that the discipline is pivoting from describing coastal ecology toward deliberately managing the planet’s carbon budget.

At the heart of the paper is a technical tour of the ocean’s “pumps.” The solubility pump begins at the air–sea interface, where carbon dioxide dissolves into seawater as dissolved inorganic carbon—a chemical pool spanning CO2, carbonic acid, bicarbonate and carbonate ions. Cold deep waters, hovering between roughly minus two and four degrees Celsius, hold far more dissolved gas, and carbon carried into them can circulate for about 1,000 years before returning to the surface. Yet the solubility pump delivers only about 10 percent of the dissolved inorganic carbon reaching the deep ocean. The biological pump does the heavier lifting: phytoplankton fix carbon into organic tissue, which is grazed, packaged into fecal pellets and aggregates, and rained downward at a rate of roughly 0.04 petamoles of carbon per year, aided by the nightly vertical migrations of zooplankton and fish. Once buried in sediment, organic carbon can remain sequestered for more than 125 million years—until volcanism or uplift returns it to the atmosphere. The carbonate pump, powered by calcifying coccolithophores, foraminifera and pteropods, complicates the ledger: building calcium carbonate releases CO2 and lowers seawater pH, partly canceling the biological pump’s gains.

The most consequential rewrite concerns microbes. Long-term carbon persistence was long attributed almost entirely to oxygen-starved sediments that slow decomposition. The review assembles emerging evidence that stabilization is instead a three-way affair involving mineral-associated organic carbon—organic matter that adsorbs onto clay particles and metal oxides—microbial transformation, and the accumulation of microbial necromass, the dead cells and residues of bacteria and archaea that resist further decay. In the open ocean, the microbial carbon pump converts labile dissolved organic carbon into recalcitrant dissolved organic carbon, a reservoir that can persist for centuries to millennia and forms through direct microbial release, viral lysis of cells and the degradation of particles. Photosynthetic marine microbes, which generate an estimated half of the oxygen on Earth, thus act simultaneously as carbon’s undertakers and its archivists. Under alkaline conditions, bacteria can even induce carbonate precipitation, potentially helping recalcitrant carbon sink into sediments—an idea the authors flag as promising but immature. Recent work synthesized in the review shows that mineral association and microbial processing jointly prolong carbon turnover in coastal wetlands, with salt marshes exhibiting exceptionally long soil carbon residence times, demolishing the old assumption that anoxia alone explains blue carbon’s durability.

Among the established vaults, the numbers remain staggering. Mangroves, covering about 14.5 million hectares—only 1.5 percent of tropical and subtropical coastlines—hold an average of 693 metric tons of carbon per hectare across their full ecosystems, roughly three-quarters of it belowground in soils that can reach extraordinary depths; peat deposits in Mexico’s Yucatán Peninsula exceed 2,700 tons per hectare. Globally, mangroves store between 5.2 and 8.6 petagrams of carbon, more than any other coastal habitat, and deliver 10 to 15 percent of coastal carbon sequestration. Salt marshes, spanning about 5.3 million hectares on every continent except Antarctica, average 287 tons per hectare and collectively hold 1.7 to 2.0 petagrams, with accumulation rates that vary widely across climate zones and sediment regimes. Seagrass meadows received the sharpest revision: a new global synthesis pegs sediment stocks at 37.7 tons per hectare in the top 30 centimeters—substantially lower than earlier figures of 165.6 tons per hectare for the top meter, which the review attributes to historical sampling biased toward carbon-rich sites. Even so, seagrasses lock away roughly one petagram of carbon across just 26.7 million hectares, less than 0.2 percent of the ocean floor, while tolerating depths of up to 40 meters.

The review’s boldest section concerns the outsiders. Macroalgae, including kelp, were long excluded because they anchor to rock, lack roots and build no sediments of their own. Yet kelp forests export an estimated 80 percent of their production as detritus and dissolved carbon to neighboring soft sediments and the deep sea, where it can be buried, and isotopic work shows canopy kelps rapidly fix carbon and leak dissolved organic matter around the clock. The sticking points are attribution—tracing exported carbon to its source after long-distance transport—and additionality, whether management truly increases burial. Shelled organisms present a different paradox: oyster and mussel shells are about 12 percent carbon locked in calcium carbonate, and reef-forming bivalves act as ecosystem engineers, slowing near-bottom currents and enhancing deposition of organic-rich biodeposits; one recent study found intensive oyster farming increased sediment carbon burial over decades. Coral reefs anchor a fierce sink-versus-source debate because calcification releases CO2 and their sediments contain less than 1 percent organic carbon. Yet the review stresses that reefs buffer waves for adjacent seagrass and mangrove ecosystems, that seagrasses in turn cut coral pathogen loads and raise seawater pH, and that cold-water coral mounds are emerging as sinks accumulating carbon faster than the surrounding seafloor. To qualify as blue carbon under influential criteria, an ecosystem must remove significant greenhouse gases, store carbon long-term, face human threats and be manageable without harm—boxes that kelp, bivalves and reefs may yet tick.

The threat ledger is grim. An estimated two-thirds of the world’s seagrass habitat has been lost; mangroves are vanishing at 0.16 to 0.39 percent annually, exceeding 8 percent in parts of Southeast Asia, where 44,485 hectares were cleared between 2000 and 2016; 1,453 square kilometers of salt marsh disappeared between 2000 and 2019; and roughly 59 percent of coral reef cover has been lost or severely degraded. Disturbance converts vaults back into chimneys: erosion re-emits about 75 percent of the carbon it exposes, degraded marsh sediments alone release an estimated 63 gigagrams of carbon per year, and remineralization tied to deforestation and land-use change accounts for 8 to 20 percent of global greenhouse gas emissions. Under high sea-level-rise scenarios, up to 30 percent of coastal wetlands could drown by 2100, squeezed between rising water and immovable infrastructure, and continued mangrove loss could ultimately release more than 3,000 teragrams of CO2. The upside is equally quantified: protecting remaining vegetated coastal ecosystems would avoid 304 teragrams of CO2 emissions each year, while restoring lost habitats could capture an additional 841 teragrams annually—while also blunting storm surges, stabilizing shorelines and supporting fisheries. Brazilian mangroves, which sequester carbon faster than most, are singled out as restoration hotspots.

The authors also weigh the lure of marine geoengineering and find it wanting, for now. Ocean iron fertilization stimulates phytoplankton blooms, but experiments show much of the carbon is remineralized before it can sink; artificial upwelling risks hauling deep, carbon-rich water to the surface and becoming a source rather than a sink; direct injection of CO2 into waters 1,000 to 3,000 meters deep carries formidable costs, leakage risks and gradual re-release over time; and alkalinity enhancement with minerals such as olivine remains experimental, with unresolved ecological consequences. The bottom line, the researchers argue, is that no technological fix yet matches the efficiency of defending what already exists—and the definition of blue carbon should widen. Some researchers now call for the term to embrace “all forms of marine, intertidal and estuarine carbon,” a change that would pull kelp, shellfish and reefs into climate policy and unlock new avenues for conservation and restoration. Given how much carbon the ocean already hides, the review suggests, the cheapest climate technology on Earth may be a mangrove root, a kelp frond and a very patient oyster.