On March 10, a journal called Science of Climate Change published a paper calling into question the foundations of climate change. The paper concluded that after accounting for some sources of uncertainty in the climate data, the ‘correct’ changes in the oceans’ heat content and Earth’s energy imbalance are practically zero. In other words, the oceans are not warming, Earth’s surface is not accumulating heat, and global warming is not happening.
The paper is more sophisticated than it seems at first sight and makes three claims with different levels of merit.
There is value in addressing them in detail because doing so reveals how we know that climate science is credible.
Heat, maths, Argo
The paper’s foremost claim is that temperature is an intensive property — meaning its value does not depend on the mass of the material — and thus scientists cannot average it in a meaningful way when estimating the amount of heat oceans hold.
Scientists have already addressed this claim. First, by the same logic, we cannot measure average air temperature, average atmospheric pressure, average sea level rise, and so on.
Second, and more importantly, scientists do not just measure and average the temperatures of different water bodies to determine the heat content. They also calculate the thermal energy. Temperature is nothing but the average kinetic energy of the atoms or molecules in a body. And thermal energy is the total kinetic energy of the water molecules. This is an extensive quantity — it depends on the number of molecules — and can be averaged. Its value has been increasing over time as well, and it also clarifies that the way scientists are handling the temperature data is correct.
Next, the paper takes issue with the Argo floats data. According to the U.S. National Oceanic and Atmospheric Administration, “Argo is an international program that collects information from inside the ocean using free drifting [devices called] profiling floats. These floats drift with the ocean currents and move up and down between the surface and a mid-water level. The floats are distributed over the global ocean to measure temperature and salinity in the upper 2,000 m.”
According to the paper, the data the floats collect have some gaps, which has created uncertainties in the final processed data that have gone underreported. To the paper’s credit, scientists have already raised and addressed the numbers it has cited on phenomena called mesoscale aliasing and deep ocean ignorance in the research literature. However, the paper’s authors inflate these uncertainties and add them up in unscientific ways. For instance, some of the errors they add up have the same underlying cause, so adding them as if they were separate errors ends up counting the same causes more than once.
To avoid this pitfall, oceanographers run the whole calculation in different ways and check whether they keep getting roughly the same answer. They check whether the calculation predicts temperatures at locations where they actually have measurements to compare with — and whether the overall estimate holds up when they remove data from the calculations. This way, they make sure their methods are robust and don’t over-count uncertainties.
Finally, if scientists have an independent estimate of total sea level rise (from altimetry satellites that use radar, say) and an independent estimate of how much new water has been added (from the GRACE satellites that use gravity), they can estimate how much the ocean has ‘expanded’. Then they compare this figure with Argo data about the ocean’s heat content. If the two match — as they do — it would mean Argo arrived at the same result as altimetry and GRACE satellites but from completely different starting points.
Balancing and filling
Finally, the paper says the CERES-Argo cross-calibration is “circular”. This is probably the most rhetorically effective piece for ‘climate doubters’ because it genuinely sounds damning. However, that is really because it misrepresents what a particular ‘adjustment’ does.
CERES is a suite of scientific instruments in Earth orbit operated by NASA. Its name stands for ‘Clouds and the Earth’s Radiant Energy System’. The instruments measure incoming solar radiation and outgoing shortwave radiation (which includes visible light) and longwave radiation (mostly heat) at the top of the atmosphere. By subtracting the incoming rate of energy from the outgoing rate, scientists can say how much heat is being ‘left behind’ in the planet’s atmosphere and on the surface.
Now, the CERES instruments have been calibrated such that they are accurate to around 1% for shortwave radiation and 0.75% for longwave. This implies an absolute uncertainty of roughly 2 W/m2 in the net energy flux. The paper says the raw CERES uncertainty is around 3-5 W/m2, which is slightly inflated in favour of the paper’s claims.
To address this, a process called EBAF — short for ‘Energy Balanced And Filled’ — makes a one-time change to shortwave and longwave fluxes at the top of the atmosphere to ensure the global mean net flux for July 2005 to June 2015 is consistent with the value measured by Argo: 0.71 W/m2.
‘Balancing’ and ‘Filling’ are separate adjustments. CERES instruments can’t see through clouds, so ‘Filling’ patches gaps in the map where data is missing. However, the paper treats the whole EBAF product as if it is just the adjustment for calibration, i.e. ‘Balancing’.
Nonetheless, as things stand, the paper’s circularity argument is partly correct: CERES data is ‘corrected’ using Argo data while Argo-derived estimates of the oceans’ heat content are validated using CERES data.
Devil in the difference
But then the paper goes wrong. The ‘Balance’ of EBAF only adjusts the mean energy flux. It does not interact in any way with the increases and drops in the temperature data over time. That comes from the CERES instruments’ raw data and is what contains evidence of the warming trend.
Specifically, the CERES instruments continuously measure the radiation coming from the Sun into the atmosphere and the outgoing longwave and shortwave radiation at the top of the atmosphere. Then computers produce a monthly global mean net flux value for every month from March 2000 to the present. These raw monthly data are off from the 0.7 W/m2 measured by Argo, so EBAF ‘adds’ or ‘subtracts’ some flux from every monthly value in the record to bring the long-period mean in line with the Argo estimate.
The monthly values after this adjustment are thus higher or lower than the raw values by exactly the same amount in every single month. This means the difference between any two months — March 2005 and March 2015, say — is unaffected by EBAF. For example, if EBAF adds 3.6 W/m2 to the measured values, the difference between 4 and 5 in the raw data is the same as the difference between 7.6 and 8.6 in the processed data, which is 1 W/m2.
The idea that Earth’s energy imbalance has been increasing over the satellite record is based on these differences, which Argo data is unconcerned with. As a result, the circularity objection proves less than what the paper has claimed it does.
Bedrock of credibility
For added measure, scientists have also estimated Earth’s energy imbalance (outgoing minus incoming) using atmospheric reanalyses, deep ocean temperature records from research vessels, and physical models informed by observed surface warming — all of which have been consistent with the CERES-Argo figures. If the imbalance were actually zero, the independent estimates would all have to be wrong for independent reasons. And the odds of that are extremely low.
In fact, the answer to whether the paper could be right is not that one of its authors is a “clarinet instructor” (which he was at the time he worked on the paper), that its rhetoric is often obscurantist, that some of its other authors have done questionable things in the past or even that it has not passed peer-review by a “prestigious” journal. It is that the paper does not perform any independent tests of the data the way credible studies do.
That is in fact the bedrock of the credibility of climate science as a whole. And any efforts that claim to overturn that must also convincingly explain how independent checks arrived at the same result while being flawed in independent ways.
mukunth.v@thehindu.co.in
Published – March 24, 2026 07:15 am IST
