Over 90 percent of the planet's trapped heat is in the ocean. How we measure ocean heat content, from expendable probes to 4,000 drifting floats — and why it matters.
If you want one number that summarises global warming, do not look at the air temperature. Look at ocean heat content — the total thermal energy stored in the sea. Because water packs away vastly more heat than air, and because the ocean covers two thirds of the planet, more than 90 percent of the extra energy trapped by greenhouse gases has gone into the ocean, with the atmosphere, melting ice and warming land sharing the remainder. Ocean heat content is therefore the most complete thermodynamic record of the warming planet: steady, difficult to fake, and brutally unambiguous. Every year for decades the upper ocean has set new high-temperature records, and the total-heat curve has climbed with a persistence that year-to-year weather cannot explain. This is also, not coincidentally, the most consequential number for kelp forests: marine heatwaves ride on top of this record, and the whole modern kelp story is what happens when a febrile ocean meets cool-water forest.
Ocean heat content (OHC) is measured in joules — the same unit as the energy to do work — and the trends are so large that scientists count them in zettajoules (10²¹ joules). The upper ocean's heat increase over recent decades adds up to the equivalent of detonating, relentlessly, several atomic-bomb energies per second, every second, for years; or to boiling more water than the oceans themselves contain, in aggregate-energy terms. Esoteric comparisons aside, the measurable, mundane fact is that a few tenths of a joule added per kilogram of a billion cubic kilometres of seawater composes, over decades, into a trend far beyond any natural variability.
Three properties make OHC a uniquely honest climate metric:
The journey from bucket thermometers to a global observing system is one of oceanography's great engineering stories.
Early oceanographers dropped insulated water bottles and reversing thermometers from ships; merchant vessels later hauled buckets of surface water whose temperature was read on deck — with all the biases buckets and engine intakes imply. From the 1960s, the workhorse became the expendable bathythermograph (XBT), a probe dropped from moving ships that radioed back a temperature profile to around 450–700 metres before being discarded. XBT archives built the first quasi-global ocean temperature records — but with uneven coverage dominated by shipping lanes, and with calibration quirks that took decades to scrub out of the long-term record.
The modern OHC record rests on Argo, one of humanity's great collaborative scientific instruments: an international array of several thousand autonomous profiling floats, drifting with deep-ocean currents, that every ten days rise from 1,000 metres "parked" depth through 2,000 metres of water to the surface, measuring temperature and salinity continuously, then transmit by satellite. At full operation the array has a float roughly every 300 kilometres across the global ice-free ocean — which is to say we now monitor the ocean's interior almost as well as we monitor the atmosphere.
Argo curves show an uninterrupted, accelerating rise in global OHC, with the hopelessly land+ice+atmosphere share left as a rounding error. The array is now expanding into BGC-Argo floats (adding oxygen, pH, nitrate and chlorophyll) and Deep Argo (profiling to 6,000 metres) — the same float family that is revolutionising our view of ocean deoxygenation and acidification.
A third, independent line of evidence comes from space. Warming water expands; thermal expansion is a major contribution to the observed rise in global mean sea level. Because satellite altimeters measure sea level globally and gravimetry missions measure the extra water mass dumped from land ice, the difference — the "expansion" component — is itself a measurement of ocean heat gain, one that integrates all depths at once, below even Argo's reach. The altimetric heat record corroborates the float record, which is exactly why the OHC number stands as solidly as it does.
The composite dataset, increasingly calibrated across instruments, adds up to a clear picture:
Two climatically famous consequences drop directly out:
When the tropical Pacific trade winds slacken, warm water piled in the west sloshes east and the ocean releases heat to the atmosphere — global mean air temperature jumps measurably in El Niño years. La Niña does the reverse. The El Niño cycle is, in energy terms, the ocean and atmosphere trading joules back and forth: atmospheric temperature wobbles on this exchange, but ocean heat content climbs right through it. This is exactly why the air-temperature record is a noisy proxy and OHC the steady truth — and why kelp coasts are so sensitive to the cycle, since El Niño's warm pool reorganises upwelling and currents across entire basins.
Every MHW event — the "Blob," the Tasman Sea events, the 2023–24 global anomalies — is laid over a heat-content baseline that has been rising for decades. What would once have been a seasonal warm spike is now a category-defining event; what would once have been a severe heatwave is now record-shattering. The OHC curve does not predict any given event, but it explains, completely, why the events keep outdoing themselves.
For a kelp forest, OHC is not an abstraction — it is the energy reservoir from which every heatwave is drawn, and the growing thermal background against which a cool-water species must survive:
That is not a counsel of despair — it is the engineering brief. Every Tidewright siting decision (fictional, but grounded in exactly this physics) starts with the local heat content and stratification trend from our observing partners, and our Holdfast hatcheries screen seedstock across the temperature range that trend implies, not the range a site saw in the 1990s. Our Canopy Ledger platform reports canopy health alongside local heat anomalies precisely because a forest's growth curve only makes sense against the thermal backdrop.
Scientific integrity matters here, so the uncertainties deserve naming:
None of this shakes the direction. Every independent dataset — floats, altimetry, hydrography, even the physics of sea-level rise itself — agrees that the ocean is accumulating heat, faster than at any point in the instrumental or proxy record. The detail is uncertain; the story is not.
A degree means different amounts of energy for air and water; seawater's enormous heat capacity means the ocean absorbs immense energy for a small temperature change. Joules measure the physics directly and are comparable across depths, regions and datasets — they're the only honest currency for a whole-ocean budget.
The target design array is about 3,000 active floats at any time; in practice the array has operated near or above that target for over a decade, maintained by more than two dozen nations deploying replacements as floats die after their typical multi-year lives.
No — the "atomic bombs per second" comparisons measure the energy of the trend relative to human-familiar energy sources, not any literal heating to boiling. Even the record total heat gain translates to average whole-ocean temperature rises of small fractions of a degree — which is precisely why the trend is invisible to bathers yet unignorable in its consequences.
No. The upper few hundred metres warm fastest; warming attenuates with depth, and the deep and abyssal ocean is only beginning its own slow response. The deep ocean is also the sparsely sampled frontier — onward Deep Argo expansion is closing that gap.
Marine heatwaves are weeks of anomalously hot ocean that can deforest entire coastlines. Here's the science behind them and what restoration can do.
Dr. Lena Fischer ·
Every few years the Pacific reorganises itself and kelp forests pay the price. How El Niño starves kelp — and how La Niña's cool afterwash lets it recover.
Dr. Mara Ellison ·
The ocean is losing oxygen — measurably and globally. How warming and nutrient stress squeeze marine life's air supply, and what that does to kelp forests.
Dr. Lena Fischer ·