Carbon dioxide accumulated in the atmosphere again in 2025, even though its annual increase slowed. The World Meteorological Organization’s latest assessment shows why a smaller rise should not be mistaken for a reversal of climate change.
Released on 27 October, the WMO Greenhouse Gas Bulletin puts the globally averaged surface concentration of carbon dioxide at 426.8 parts per million. Methane and nitrous oxide also reached record concentrations in the agency’s observational record.
The news is about what remained in the air, rather than how much countries emitted during a single year. Understanding that distinction is essential to judging both the warning and the response it calls for.
Three gases, three new highs
WMO reports carbon dioxide at 53% above its pre-industrial level before 1750. Methane reached 1,948 parts per billion, an increase of 167% over that baseline. Nitrous oxide reached 339.1 parts per billion, 26% above the baseline.
Those percentages need careful reading: 153% of a previous level means 53% above it. The concentrations also use different units. One part per million is a thousand parts per billion, so the three headline numbers cannot be compared as though they share a scale.
The bulletin gives uncertainties of ±0.2 parts per million for carbon dioxide, ±2 parts per billion for methane and ±0.1 parts per billion for nitrous oxide. These are measurements with stated precision, not exact counts of every molecule.
A slower rise still adds to the total
In its accompanying assessment, WMO says carbon dioxide rose by 2.8 parts per million between the annual averages for 2024 and 2025. That was below the exceptional previous increase, but still a substantial addition.
Emissions are the flow entering the atmosphere; concentration describes the accumulating stock. Land ecosystems and oceans absorb part of the carbon dioxide people release. The amount left in the air therefore depends on both emissions and the changing performance of those natural sinks.
WMO attributes the slower growth partly to recovering land uptake and reduced fire emissions after the 2023–2024 El Niño. It warns that the powerful El Niño developing in 2026 could again put pressure on land carbon sinks. This is a warning about conditions affecting the carbon balance, not a measured annual total for 2026.
What the observations can tell us
A complementary measure comes from the US National Oceanic and Atmospheric Administration’s Annual Greenhouse Gas Index. It estimates that the heating influence of long-lived greenhouse gases was 56% greater in 2025 than in 1990.
That does not mean global temperature rose by 56%. The index measures a change in the planet’s energy balance caused by those gases. It is also not a forecast of a particular flood, drought or heatwave.
NOAA builds its record from air samples collected across a global network. Its marine boundary-layer average and WMO’s global surface average are different products; their concentration figures should not be substituted for one another without checking methods. Both records show the continuing build-up.
Global monitoring also cannot identify every source that needs fixing. SMD’s reporting on methane measurements in Indianapolis illustrates the next step: combining observations and modelling to distinguish a major landfill source from emissions distributed across a city.
Persistence does not make action pointless
Carbon dioxide’s long influence means that stopping further warming and quickly reversing past warming are different ambitions. The carbon-cycle assessment in Canada’s Changing Climate Report explains that warming from carbon dioxide is closely related to cumulative emissions.
Reaching net-zero carbon dioxide emissions, alongside strong reductions in other greenhouse gases, is therefore central to stabilising global temperature. A long atmospheric legacy does not mean temperatures must keep climbing at the same rate regardless of future decisions.
For readers confronting apparently contradictory headlines, the useful questions are straightforward. Is a number describing emissions, atmospheric concentration or heating influence? Is the comparison with last year, 1990 or the pre-industrial period? Does “slower” mean an actual decline, or simply a smaller increase?
SMD’s guide to checking climate claims against the evidence offers further context for separating short-term variability from long-term change.
The new bulletin’s message survives those distinctions: the atmospheric burden rose again. Progress will require sustained reductions at the sources, with measurements continuing to test whether those efforts are changing what accumulates overhead.
Featured photograph: Mauna Loa Observatory, Hawaii, 18 June 2017. Illustrative archive photograph, not a photograph of the 2025 measurements or the bulletin launch. Photo: Ivtorov, original image, CC BY-SA 4.0. Resized for web display; photograph adaptations remain under the same licence.
