Climate change
How we know the planet's temperature has changed before, how we know this time is different, and what the ice and the thermometers actually show.
Last updated September 22, 2026
Reading temperature out of ice and rock
Nobody had a thermometer 500,000 years ago, so reconstructing that far back relies on proxies — physical traces that change in a known way with temperature. The most direct is the ratio of heavy to light isotopes of hydrogen and oxygen (deuterium/hydrogen, oxygen-18/oxygen-16) locked into ancient ice and ocean sediment. Water molecules carrying the heavier isotope condense out of a cloud and fall as snow more easily in cold air than warm air, so the isotope ratio in each year's snow layer is a direct fingerprint of the temperature when it fell. Drill a two-mile-deep ice core through Antarctica and you're pulling up 800,000 annual snapshots, oldest at the bottom.
800,000 years of natural cycles
Temperature difference from the last-1,000-years average, from the EPICA Dome C deuterium record — Jouzel et al. 2007, Science; NOAA/WDS Paleoclimatology.
The record oscillates roughly every 100,000 years, driven by slow, predictable wobbles in Earth's orbit and tilt (Milankovitch cycles) that change how much summer sunlight reaches the high latitudes. Each full swing — ice age to interglacial and back — plays out over 10,000 to 20,000 years. That pace is the baseline the next chart is measured against.
The last 145 years, measured directly
Land-ocean temperature index, annual mean, vs. 1951–1980 — NASA GISS Surface Temperature Analysis (GISTEMP v4).
Once thermometers, ships, and eventually satellites could measure temperature directly, isotopes stopped being necessary — but the two records describe the same planet, so they're comparable. The rise here, roughly 1.3°C since 1880, is similar in size to a full natural glacial-to-interglacial swing, but it happened in under 150 years instead of 10,000–20,000. NASA, NOAA, and the Intergovernmental Panel on Climate Change (IPCC) all attribute the pace of this particular rise to the buildup of greenhouse gases from human activity, primarily the burning of fossil fuels.
Nonpartisan, plainly
We don't take a position on what policy should follow from any of this — carbon pricing, energy permitting, or adaptation spending. None of the above are disagreements with legitimate arguments on one or more sides, and it's not this page's job to settle them. What isn't a matter of opinion is what the ice and the thermometers measure: these are the sourced numbers, plainly.
Talking points
These are the questions we think you should ask those who are running for office and will represent you. We don't give our opinion on the answer, but we DO think you should be talking about them.
- Should the U.S. set a binding national target for greenhouse gas emissions, and who should enforce it?
- Should climate research funding be protected from being cut based on which party controls Congress?
- Should local governments be required to plan infrastructure funding around measured climate trends, like sea-level rise and extreme heat?
Read more
- NOAA/WDS Paleoclimatology — EPICA Dome C 800KYr Deuterium Datancei.noaa.gov
The ice-core isotope dataset behind the 800,000-year chart above (Jouzel et al. 2007).
- NASA GISS Surface Temperature Analysis (GISTEMP v4)data.giss.nasa.gov
The instrumental global temperature record behind the modern chart above.
- NOAA: How do scientists study ice cores?climate.gov
A plain-language explainer on the isotope method used to reconstruct ancient temperature.
- IPCC Sixth Assessment Report — Synthesis Reportipcc.ch
The UN climate science body's periodic assessment of the physical evidence and its causes.