You asked
Is climate change real?
That isn't the same as proof. Nothing here is. Below is how much testing that idea has survived, what is still argued about, and how to go and check it yourself.
Most of the evidence points toward it. The needle can still move, and it has moved before — that is the point of keeping it on a dial rather than stamping a verdict on it.
Where the needle sits is worked out from the evidence further down this page — how much points each way, weighted by how strong each piece is. It is not a measure of how certain anyone feels.
Why they think that
Four different teams in different countries measure the world's temperature separately, using different methods. They all land in the same place: roughly 1.3°C warmer than the 1800s. When rivals using different methods keep getting the same answer, that answer is hard to dismiss.
Carbon from burnt coal and oil is slightly different, chemically, from carbon that was already moving around in nature. The extra carbon that has built up in the air matches the burnt kind. It is a bit like a receipt.
If the sun were simply putting out more heat, the whole atmosphere would warm together. Instead the upper atmosphere is getting colder while the surface warms. That is the fingerprint of heat being trapped near the ground — not of a brighter sun.
What is still genuinely argued about: not whether it is happening, but how much warming we get for a given amount of carbon, how bad the effects will be, how fast, and what is worth doing about it. Those are real arguments among serious people. Whether it is happening is not one of them.
How sure can anyone be
Here is the bit that often gets lost. Science does not work by proving an idea true. Someone has an idea, and then everybody else spends years trying to prove it wrong.
If lots of people attack an idea from lots of angles and it keeps standing up, we trust it more. Not because it has been proved — because it has survived a beating. That is all anyone means by “settled science”: an idea that has taken a great many hits and has not fallen over.
This one has taken about 150 years of hits, from people who very much wanted to knock it down, and it is still standing. That is why the needle sits where it does.
So we will never tell you something is a fact. We will tell you how much testing an idea has survived and which way the evidence leans. Those can change — and when they do, this page changes, and we keep the old version so you can see what we used to say.
And we could simply be wrong. We might have missed a study, weighed something badly, or written it up carelessly. That is exactly why the next section exists.
Do not take our word for it
Trusting an expert is a shortcut. Shortcuts are fine most of the time — but you should not have to take one, and you should not have to trust us either. Everything below is free and needs no training.
NASA and the Met Office publish their raw temperature records online for anyone. You can plot them yourself. If someone tells you the world is not warming, this takes two minutes to check.
Put their name into Google Scholar. It is free. Have they published research on this subject, or are they a qualified expert in something else being quoted about climate? Both happen a lot, in both directions.
News stories about research usually link to it, or name the journal. Read the abstract — the short summary at the top. Headlines routinely say more than the study does.
One study is a hint. Five separate teams finding the same thing is a pattern. A result nobody has managed to repeat is interesting but not yet reliable — that is true of results you like as well as ones you do not.
“Linked to” is not “causes”. “May” is not “does”. “Doubles your risk” means nothing until you know what the risk was to start with — double a tiny number is still a tiny number.
What to watch out for
“97% of scientists” gets quoted constantly. But one of those studies counted papers, another surveyed people, and they asked different questions. The number is solid; the shorthand is sloppy — and that sloppiness is what critics attack, fairly.
They need different evidence and carry very different confidence. Coverage on both sides blurs them, in opposite directions — one to make it feel urgent, the other to make it feel overblown.
In either direction. People who actually work on this talk in ranges and probabilities, because that is what the evidence supports. Total confidence is usually a sign someone is selling something.
Going deeper
Weighted by how strong each piece is — not by how many times it gets repeated. One large review of many studies outweighs three small ones.
Four independent global temperature records, kept by separate institutions using different methods, agree that 2015–2024 was the warmest decade measured. s3 s4
The stratosphere is cooling while the surface warms. Greenhouse gases produce that pattern; a brighter sun would not. s1
How much warming results from doubling CO₂ is put at 2.5–4°C. That range is wide enough to matter enormously for policy, and narrowing it is active research. s1
Tying any single storm or heatwave to climate change is a matter of probability, not proof. Coverage saying one event was “caused by” it goes further than the science does. s1 s5
The 1998–2012 “pause” was widely cited as proof the models were wrong. It was later explained by heat moving into the oceans and gaps in Arctic measurement, and warming since 2013 has been faster than before it. Worth knowing this argument was made, tested, and did not survive. s4
| What was counted | How | When | Result |
|---|---|---|---|
| 11,944 paper summaries | Those taking a position | 2013 | 97.1% |
| 2,780 earth scientists | Survey of people | 2021 | 98.7% |
| 88,125 papers | Literature analysis | 2021 | 99% |
The serious sceptical case is not that warming is not happening. It is that the climate may be less sensitive to carbon than the middle estimates assume, that adapting is undervalued next to cutting emissions, and that the economic damage figures behind policy are as uncertain as anything in the physical science.
Why most specialists do not find it enough: The low-sensitivity argument rests on a subset of estimates that struggle to explain past climate shifts, while several independent lines of evidence converge near 3°C. The economic half lands better — that uncertainty is real and openly acknowledged. But it is an argument about what to do, not about whether it is happening.
It covers whether warming is happening and what is causing it. It does not judge which policies are worth their cost — that is a question about values and trade-offs, and evidence alone does not settle it. Forecasts for particular regions, and the timing of tipping points, are far less certain than the cause is.
We would much rather know. No account, no sign-in, and we don’t ask who you are — a site that tells you not to trust experts can hardly make you identify yourself to say we’re wrong.
Does this match the evidence as you understand it?
Did both sides get a fair hearing?