Is Climate Change Real?
Science vs. Media Coverage

Climate skeptics often argue that temperature records are unreliable, manipulated by urban heat islands, or limited to a small number of politically motivated institutions. This is testable. Four independent teams — at NASA (U.S.), NOAA (U.S.), the Met Office/CRU (UK), and Berkeley Earth (independent nonprofit) — assembled global temperature records using different station networks, methodologies, and quality-control procedures.

The result: all four datasets show essentially the same warming signal, agreeing within approximately 0.1 °C at any given year. Independent replication across different methodologies is the strongest possible evidence of a real signal.

Source: IPCC AR6 WGI, Table SPM.1. Berkeley Earth uses ~40,000 station records vs. ~7,000 in earlier NOAA analysis. The agreement across sample sizes and methodologies rules out systematic station-bias explanations.

2024 became the first calendar year to exceed 1.5 °C above pre-industrial levels (the Paris Agreement threshold), according to Berkeley Earth, NOAA, and Copernicus Climate Change Service data. 2025 ranked as the 3rd-warmest year on record. All ten of the hottest years in the instrumental record have occurred since 2010.

One of the most persistent skeptic arguments is that warming is an artefact of cities growing around weather stations. Berkeley Earth specifically tested this: their analysis of rural-only stations shows the same warming trend as urban stations. Satellite datasets (RSS and UAH lower troposphere) — which have no surface station bias whatsoever — also show warming, though with somewhat different short-term variability. The urban heat island effect is real and scientifically documented, but it affects local temperature in cities, not the global average trend.

The longest continuous direct atmospheric CO₂ measurement is the Keeling Curve, maintained at NOAA's Mauna Loa Observatory in Hawaii since 1958. The site was chosen specifically for its remoteness from industrial sources. The measurement has been independently replicated at dozens of stations globally.

The physics of the greenhouse effect has been understood since the 19th century (Fourier 1824, Tyndall 1859, Arrhenius 1896). CO₂ absorbs outgoing infrared radiation and re-emits it in all directions, trapping heat in the lower atmosphere. This is not a hypothesis — it is measured directly in laboratory spectroscopy and confirmed by satellite observations of outgoing longwave radiation.

The expected physical fingerprint of greenhouse-gas-driven warming includes: (1) greater warming at night than during the day — confirmed; (2) greater warming at the poles than the tropics — confirmed; (3) warming of the troposphere coupled with cooling of the stratosphere — confirmed; (4) increased downward longwave radiation at the surface — directly measured. Natural solar forcing would produce a different fingerprint: uniform tropospheric warming without stratospheric cooling.

Critics correctly note that CO₂ has varied naturally over Earth's history — between roughly 180 ppm (ice ages) and 280 ppm (interglacials) over the past 800,000 years based on Antarctic ice cores. What makes the current increase anomalous is its speed: those natural cycles operated over 10,000–100,000 years. The current 150-ppm increase above the pre-industrial baseline has occurred in approximately 200 years — orders of magnitude faster than natural glacial-interglacial transitions.

Surface air temperature records cover only the atmosphere, which holds a small fraction of Earth's total heat. The ocean absorbs over 90% of the excess heat trapped by greenhouse gases. Ocean heat content (OHC) is therefore a more reliable measure of planetary energy imbalance than surface temperatures alone — it is less affected by short-term atmospheric variability like El Niño events.

Source: NOAA National Centers for Environmental Information (NCEI), Ocean Heat Content dataset. Independent analyses from Cheng et al. (2023) using Argo float network data broadly concur.

Global mean sea level is measured by satellite altimetry (TOPEX/Poseidon from 1993, followed by Jason-1/2/3, and Sentinel-6) and by a global network of tide gauges dating back to the 19th century. The two measurement systems are independent and agree closely on trends.

The acceleration of sea level rise matters as much as the absolute amount. A constant rate would suggest a manageable linear future. Acceleration means the rate itself is increasing — consistent with ice sheet dynamics adding to thermal expansion, and inconsistent with a stable or natural cycle explanation.

The National Snow and Ice Data Center (NSIDC) has tracked Arctic sea ice extent using satellite passive microwave data continuously since 1979. This is a 46-year dataset with no surface station issues — the data comes from orbiting satellites measuring microwave emissions from ice and open ocean.

The World Glacier Monitoring Service (WGMS) maintains records from over 150 reference glaciers worldwide, with some measurements dating to the 1880s. Mass balance (the difference between snow accumulation and melt) has been negative globally since the 1980s and accelerating since the 2000s.

Note: m w.e. = metres water equivalent. Independent confirmation from NASA GRACE and GRACE-FO satellite gravimetry (which measures changes in Earth's gravitational field caused by mass redistribution) matches WGMS direct measurements.

Ice sheets are a different category from mountain glaciers — they are continental-scale bodies of ice that store most of the world's freshwater. Both Greenland and Antarctica are losing mass, confirmed by three independent methods: GRACE satellite gravimetry (mass change), altimetry (surface height change), and input-output mass balance (measuring snowfall in vs. ice flow out). All three agree that both ice sheets are losing mass at accelerating rates.

Attribution science answers the question: of the observed warming, how much is caused by which forcing? IPCC AR6 (2021) produced the most comprehensive attribution analysis to date, drawing on thousands of peer-reviewed papers.

Multiple independent physical fingerprints distinguish greenhouse warming from natural forcing:

Solar output has been directly measured by satellite since 1978 (ACRIM and later SORCE datasets). Since 1980, solar irradiance has shown no statistically significant upward trend — in fact, the period from approximately 2000–2020 had slightly lower solar output. Observed warming has continued and accelerated during this period of flat or declining solar forcing. The correlation between solar output and global temperature that existed in the early 20th century broke down after 1980, precisely when CO₂ forcing overtook solar as the dominant driver.

Core scientific conclusions — that warming is occurring, that it is human-caused, and that it will continue — are well-represented in mainstream media. The consensus itself is not overstated: multiple surveys of climate scientists (e.g., Cook et al. 2013, Lynas et al. 2021) find 97%+ agreement on human-caused warming among actively publishing researchers. This figure appears frequently in coverage and is defensible.

A documented pattern in media coverage — particularly in earlier decades — was “false balance”: giving equal airtime or column space to scientists representing 97% consensus and contrarian researchers representing 3%. A 2004 analysis by Boykoff and Boykoff (Global Environmental Change) found 53% of US newspaper articles gave roughly equal weight to both sides. This created a public perception of scientific controversy that did not reflect the actual distribution of expert opinion. Subsequent media guidelines have reduced but not eliminated this pattern.

Multiple peer-reviewed studies (Oreskes and Conway, Brulle 2014, Supran and Oreskes 2017) have documented concerted and funded efforts to manufacture doubt about climate science, drawing parallels to tobacco industry tactics. ExxonMobil's internal research from the 1970s and 1980s accurately modeled future warming that the company's public communications then disputed. This is documented from internal corporate records and has been reported in peer-reviewed journals (Supran et al., Science 2023).