How Do
We Know?

Multiple independent studies have found that over 97% of actively publishing climate scientists agree that recent warming is primarily caused by human activities. This is not a vote or a poll — it is the result of examining the peer-reviewed literature.

The consensus is reflected in every major scientific organization in the world: NASA, NOAA, the American Meteorological Society, the Royal Society, and the national academies of science of every G8 nation. The IPCC — which synthesizes thousands of peer-reviewed studies — states with "unequivocal" certainty that human influence has warmed the climate.

The global temperature record is built from thousands of weather stations, ocean buoys, ships, weather balloons, and satellites. Multiple independent research groups — NASA, NOAA, the UK Met Office, Berkeley Earth — analyze the raw data using different methods and reach the same conclusion: the planet is warming.

The instrumental record goes back to the mid-1800s. Before that, scientists use proxy records — tree rings, coral cores, sediment layers, and ice cores — to reconstruct past climates. These proxies confirm that the current warming is unprecedented in at least 2,000 years, and likely much longer.

Ice cores drilled from Antarctica and Greenland contain tiny air bubbles — ancient atmosphere trapped as snow compacted into ice over hundreds of thousands of years. By analyzing these bubbles, scientists can measure the exact CO₂ and methane concentrations of past atmospheres.

The data is unambiguous: CO₂ levels have never exceeded 300 ppm in 800,000 years of ice core records — until the industrial revolution. Today, CO₂ stands at over 420 ppm. The correlation between CO₂ and temperature in the ice core record is one of the most powerful pieces of evidence for the greenhouse effect.

Climate models are mathematical representations of the Earth system — atmosphere, oceans, land, and ice — built on fundamental physics: thermodynamics, fluid dynamics, radiative transfer. They are not guesses or extrapolations; they are physics simulations.

The models' reliability is demonstrated by their track record: projections made in the 1980s and 1990s have proven remarkably accurate. Models correctly predicted stratospheric cooling, polar amplification, and the pattern of warming over land versus ocean. When models are run without human greenhouse gas emissions, they cannot reproduce the observed warming — confirming human causation.

Attribution science is a relatively new field that asks: how much did climate change influence a specific weather event? Scientists run thousands of climate model simulations — some with human greenhouse gas emissions, some without — and compare the probability of the event occurring in each scenario.

This approach has produced striking results. The 2021 Pacific Northwest heat dome was "virtually impossible" without climate change. The 2019–20 Australian bushfires were made at least 30% more likely. The 2022 Pakistan floods — which submerged one-third of the country — were made significantly more intense. Attribution science is transforming how we understand the human cost of climate change.