How Do We Know Carbon Dioxide Is Rising? The Mauna Loa Record Explained

Earth & Climate · Published 2026-10-22 · Sources checked 2026-10-22

Short answer: because it has been measured directly, every hour, since 1958. Air sampled near the summit of Mauna Loa in Hawaii showed a monthly average of 426.07 parts per million (ppm) of carbon dioxide in September 2026, up from 424.37 ppm a year earlier, according to the US National Oceanic and Atmospheric Administration (NOAA). Air trapped in ice cores shows the level stayed at around 270 to 280 ppm from the end of the last ice age until about 1850. NOAA's global network puts the 2024 average at 422.80 ppm and the recent rise at about 2.6 ppm a year, faster than in the 1980s and 1990s.

This page explains how the measurement is made, why scientists trust it, what the numbers show and what they do not show. All figures come from NOAA's Global Monitoring Laboratory pages listed at the end.

Where and how the measurement started

NOAA describes the Mauna Loa carbon dioxide data as "the longest record of direct measurements of CO₂ in the atmosphere". The measurements were started by C. David Keeling of the Scripps Institution of Oceanography in March 1958, at a NOAA facility. NOAA began its own carbon dioxide measurements at the site in May 1974, and the two programmes have run in parallel since then. The combined Scripps and NOAA series is the curve widely known as the Keeling Curve.

The observatory sits near the summit of Mauna Loa at an altitude of 3,400 metres. NOAA's explanation of the method gives three reasons for confidence that the numbers reflect the global atmosphere:

• the site is placed to sample air masses that represent very large areas;
• all measurements are "rigorously and very frequently calibrated";
• independent measurements made at the same site are compared continuously, which puts the accuracy at generally better than 0.2 ppm.

What "parts per million" means here

Strictly, NOAA does not measure a "concentration". It measures the mole fraction in dry air: the number of carbon dioxide molecules per million molecules of air after the water vapour has been removed. NOAA's own example: 413 ppm means that in every million molecules of dry air there are on average 413 carbon dioxide molecules. Removing water vapour matters because humid air would otherwise dilute the count and make the numbers jump with the weather.

The monthly figures are built up carefully. Each monthly mean is an average of daily means, which are averages of hourly values, but only for hours when "background" conditions prevail; NOAA links a separate explanation of how those background hours are chosen. The most recent year of data is marked preliminary until reference gases are recalibrated and other quality checks are finished.

What the numbers show

Sources: NOAA Global Monitoring Laboratory, "Trends in Atmospheric Carbon Dioxide" (Mauna Loa monthly means, last updated 9 October 2026) and "The NOAA Annual Greenhouse Gas Index" (updated fall 2025).

The growth-rate table on NOAA's site shows how uneven individual years are. Using globally averaged marine surface data, it lists an annual increase of 0.96 ppm for 1959, 2.95 ppm for 2015 and 3.76 ppm for 2024, with 2.06 ppm listed for 2025. (NOAA's greenhouse gas index report quotes 3.72 ppm for the same 2024 increase; NOAA notes that values in the table are subject to small revisions after quality control.)

Why the yearly increase jumps around

NOAA defines a year's growth rate as the difference in carbon dioxide between the end of December and the start of January of that year. It "represents the sum of all CO₂ added to, and removed from, the atmosphere during the year by human activities and by natural processes". So a single year can be unusually high or low because oceans and plants absorbed less or more than usual, even if emissions changed little. That is why scientists look at decade averages, and those averages have been rising.

How we know the extra carbon dioxide comes from people

NOAA's greenhouse gas index report states that after about 1850 carbon dioxide "began rising in proportion to emissions resulting from fossil fuel combustion", citing the Global Carbon Budget. NOAA's isotope pages state the same conclusion directly: the increase is caused by burning coal, oil and natural gas.

One detail surprises many readers. Only about half of the carbon dioxide released by burning fuel stays in the air. The rest, NOAA explains, is absorbed mainly by the oceans and, more recently, by land ecosystems. The rising curve at Mauna Loa is therefore what is left after nature has taken up a large share.

What this has done to the heat balance

NOAA also tracks how much extra heat long-lived greenhouse gases trap, a quantity called radiative forcing. Its Annual Greenhouse Gas Index found that radiative forcing from these gases was 54% greater in 2024 than in 1990, and that carbon dioxide is "by far the largest contributor" to that increase. NOAA built the index from measured gas amounts on purpose: projections of future climate carry large uncertainties, while the measured abundances and the heating they add have little uncertainty.

What the Mauna Loa record does not tell you

• It is not a temperature record. It measures a gas. The link to warming comes from the physics of greenhouse gases and from separate temperature records, which this article does not cover.
• One site is not the whole picture. Mauna Loa is the longest direct record, but NOAA's global figures come from a network of sites; the 2024 global average (422.80 ppm) and Mauna Loa monthly values are different measures.
• The newest numbers can move a little. The latest year is preliminary, and NOAA notes that growth-rate values may be revised slightly after quality control.
• It does not forecast. The record shows what has happened. How fast carbon dioxide rises from here depends on future emissions and on how much oceans and land keep absorbing.

Common questions

Why measure on a volcano?

The observatory is at 3,400 metres, where it samples air that represents very large areas. Only hours when "background" conditions prevail are used for the averages, according to NOAA.

Does the curve go up smoothly?

No. NOAA's graphs show a seasonal cycle on top of the long-term rise, and NOAA publishes a version corrected for that average seasonal cycle. The long-term trend is clear in both.

What was the level before industry?

Air bubbles in ice cores show it was stable at around 270 to 280 ppm from the end of the last ice age until about 1850, according to NOAA's greenhouse gas index report, which cites the IPCC.

How accurate are the measurements?

NOAA reports that comparisons of independent measurements at Mauna Loa put the accuracy at generally better than 0.2 ppm.

Was 2024 a record year for the increase?

NOAA's index report calls the January 2024 to January 2025 increase of 3.72 ppm record-breaking, larger than the previous largest annual increase, in 2015.

Sources and how to check

• NOAA Global Monitoring Laboratory. "Trends in Atmospheric Carbon Dioxide (CO₂): Mauna Loa, Hawaii." gml.noaa.gov/ccgg/trends (last updated 9 October 2026).
• NOAA Global Monitoring Laboratory. "Annual Mean Global Carbon Dioxide Growth Rates." gml.noaa.gov/ccgg/trends/gl_gr.html
• NOAA Global Monitoring Laboratory. "The NOAA Annual Greenhouse Gas Index (AGGI)." Updated fall 2025. gml.noaa.gov/aggi/aggi.html
• Tans P. and Thoning K., NOAA Global Monitoring Laboratory. "How we measure background CO₂ levels on Mauna Loa." gml.noaa.gov/ccgg/about/co2_measurements.html
• NOAA Global Monitoring Laboratory. "Isotopes: The Basics — Isotopic Fingerprints." gml.noaa.gov/ccgg/isotopes
• Scripps Institution of Oceanography. "The Keeling Curve." keelingcurve.ucsd.edu
• All pages read on 22 October 2026. Mauna Loa values change monthly; check NOAA's page for the latest figure.