Ocean Heat Content: How We Measure a Warming Ocean
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.
Dr. Lena Fischer ·
Nearly 4,000 drifting robots take the ocean's temperature from within. How Argo profiling floats work, what they measure, and why the data is free.
Somewhere in the Pacific right now, a metre-long cylinder is finishing a ten-day drift in the dark, a kilometre below the waves. A small pump inflates an external bladder, the float becomes lighter than the water around it, and it begins to rise — slowly, about a hand's length per second — measuring temperature and salinity all the way up through two kilometres of ocean. At the surface it radios its data to a satellite and sinks again. That cycle, repeated by an international fleet of thousands of such floats, is Argo: the single most important ocean-observing system ever built, and the reason we can say with confidence how much heat the ocean is absorbing.
For most of oceanography's history, measuring the ocean's interior meant sending a ship. Research vessels are magnificent but scarce — a few dozen worldwide — and even routine expendable-probe lines from merchant ships covered mostly the major shipping lanes of the northern hemisphere. Huge regions, especially the Southern Ocean, went unmeasured for entire seasons. The deep ocean's temperature record was a patchwork: dense along sea lanes, empty elsewhere, frozen every southern winter.
Climate science needed something different: measurements everywhere, always — every basin, every season, storm or calm, decade after decade. In the late 1990s oceanographers proposed an audacious answer: thousands of free-drifting autonomous profilers, each cheap enough to lose, together dense enough to see the whole ocean. The first floats launched around 2000; by 2007 the array reached its design target of 3,000 active floats, a density of roughly one per 300 kilometres across the ice-free ocean. It has operated continuously ever since, contributed and governed by more than two dozen countries.
An Argo float is elegant precisely because it has almost no moving parts. There is no propeller. The float is a pressure case about 1.3 metres long containing a battery pack, a computer, a satellite modem and a buoyancy engine — a hydraulic pump that moves oil between an internal reservoir and an external bladder. Inflate the bladder and the float displaces more water and rises; deflate it and the float sinks. Small pistons in some models shift internal mass to control the angle of ascent.
Attached to the case is the sensor package: a CTD measuring conductivity (for salinity), temperature and depth — the same measurement, at miniature scale, as the shipboard CTD that built classical oceanography. The profiling sensors are astonishingly stable: temperature accurate to thousandths of a degree, held for years without servicing.
A float's life has a heartbeat of about ten days:
A float repeats this cycle for a typical lifespan of around four to five years — well over a hundred profiles — on one battery, after which it dies quietly and, being mostly aluminium inside its case, sinks. The fleet's steady state requires roughly 800–1,000 new deployments a year, launched from research vessels, ships of opportunity and even small aircraft. Sailors regard tossing an Argo float over the stern as a small honour.
The array transformed every field it touches:
The original float measures the physics only. Two expansions are under way. BGC-Argo floats add oxygen, pH, nitrate, chlorophyll fluorescence, particle backscatter and irradiance sensors — opening a window on ocean deoxygenation, acidification and plankton dynamics in exactly the regions, like the Southern Ocean, where ships rarely go. Deep Argo floats engineered for full ocean depth (down to around 6,000 metres) will do for the abyssal ocean what the core array did for the upper two kilometres, where a substantial share of the warming now resides. Together these extensions are known as OneArgo.
The expansion is hard-won: biogeochemical sensors drift in calibration over years, so the program maintains a careful delayed-mode quality-control chain, cross-checking every profile against nearby floats, shipboard reference data and climatology before certifying it.
Argo's most influential engineering decision was not mechanical. All Argo profiles are made publicly available, in real time, within about 24 hours of transmission, with higher-quality delayed-mode versions following within roughly a year. No embargo, no national gatekeeping, no proposal required. Schoolchildren catch floats surfacing on live maps; graduate students run global analyses from laptops; weather services assimilate profiles within hours.
That openness changed oceanography's culture. It made a global instrument genuinely global — the country's float in one nation's waters serves every nation's science — and it is the model for our own open data programme at Tidewright, where coastal temperature and canopy datasets are published the same way: free, versioned, citable.
Argo floats do not operate in shallow coastal water — they need depth to dive, and kelp forests live at the landward edge floats cannot reach. But the floats frame everything that happens on that edge. Kelp communities experience ocean conditions set far offshore: the basin-wide heat that becomes a heatwave, the deep reservoir that feeds upwelling, the oxygen trends that reach the shelf. When we plan a restoration season, Argo's subsurface temperature anomaly maps are as much a part of the briefing as satellite ocean colour and diver surveys.
The full ocean-observing stack is a relay: satellites photograph the surface, floats profile the open-ocean interior, gliders thread the shelf and boundary currents, and moored sensors plus divers guard the coast itself. Remove any leg and the picture tears; together they give the ocean a continuous voice.
The core array targets about 3,000 active floats in steady state, achieved in 2007. With biogeochemical and deep floats added, the active total is nearer 4,000, sustained by roughly 800–1,000 new deployments per year from more than two dozen countries.
By buoyancy control: a hydraulic pump moves oil into an external bladder, increasing the float's displaced volume until it rises; retracting the oil makes it sink. The float cannot steer — which is exactly the point, since its unsteered drift maps the deep currents.
Standard floats profile the upper 2,000 metres. The Deep Argo program is deploying floats rated to around 6,000 metres to monitor the warming abyssal ocean.
Yes. Profiles are distributed openly within about 24 hours via two global data centres, with delayed-mode, quality-controlled versions typically available within a year. Anyone — scientists, schools, companies — can download the full archive.
After roughly four to five years the battery is exhausted; the float fails at depth or drifts until it can no longer surface, then sinks. Floats are designed with minimal hazardous materials, and recovery or beaching incidents are rare given their open-ocean missions.
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.
Dr. Lena Fischer ·
Ocean gliders ride buoyancy, not propellers, for months at sea. How these slow torpedo-shaped robots survey storms, currents and kelp coasts on sips of power.
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 ·