“Scientists Predicted an Ice Age in the 1970s”: What the Literature Actually Looked Like

The useful answer is no: climate scientists in the 1970s did not reach a consensus that an ice age was imminent. Some researchers investigated cooling, particularly the cooling effects of air pollution. Newspapers and magazines sometimes amplified those possibilities. But a retrospective review of the scientific literature found substantially more papers anticipating warming than cooling.

That does not mean the cooling research was imaginary or foolish. Temperatures had cooled across part of the available record, aerosol pollution can cool the climate, and scientists were still building the observing systems and models needed to separate competing influences. The historical record is a story of incomplete data, legitimate uncertainty and rapidly improving climate physics—not a failed ice-age consensus.

TL;DR

• A review of climate papers published from 1965 through 1979 classified 44 as anticipating warming, seven as anticipating cooling and 20 as neutral or making no temperature projection. That is a literature sample, not a poll of every scientist or a census of every climate paper.
• Researchers had a real reason to study cooling: a temperature record used at the time showed about 0.3°C of cooling from roughly 1940 to 1970, although it covered 20°N to 90°N rather than the whole globe.
• A frequently cited 1971 paper calculated what could happen under a large, sustained increase in atmospheric aerosols. Its ice-age result was conditional, not a prediction that such an increase would occur.
• By 1979, a major National Research Council assessment estimated substantial warming from doubled atmospheric CO2.
• The modern conclusion that humans are warming the planet rests on observations, physical mechanisms, palaeoclimate evidence and models—not on simply extending a recent trend line.

What the published literature shows

The most direct way to evaluate the claim is to examine the research literature rather than assemble a collection of dramatic headlines. In a 2008 review, Thomas Peterson, William Connolley and John Fleck examined climate papers published from 1965 through 1979. They classified 44 as anticipating warming, seven as anticipating cooling and 20 as neutral or offering no temperature projection. The review published in the Bulletin of the American Meteorological Society therefore found no cooling consensus; papers anticipating warming outnumbered those anticipating cooling by more than six to one.

Those numbers need to be interpreted correctly. They describe the papers included under the review’s selection and classification method. They are not a poll of all scientists, a count of personal opinions or a literal census of every climate-related publication produced during those years. Even with that limitation, the distribution is difficult to reconcile with the claim that climate science as a field had settled on an imminent ice age.

Historical evidence	What it supports	What it does not establish
A mid-century cooling interval in a partly Northern Hemisphere temperature record	Scientists had an observational reason to investigate cooling influences	That the entire globe was entering a new ice age
Seven cooling papers in the 1965–1979 literature review	Some peer-reviewed research anticipated cooling	A consensus among climate scientists
A 1971 high-aerosol model experiment	Large aerosol increases could produce strong cooling in the model	A forecast that aerosol concentrations would rise by that amount
Popular articles about global cooling	Cooling scenarios attracted public and editorial attention	The balance of findings in the scientific literature
A 1979 assessment of doubled CO2	CO2-driven warming had become quantitatively assessable	That every uncertainty in climate prediction had been resolved

Why cooling was a serious research question

Climate science did not invent the cooling question out of nowhere. NASA’s historical account notes that a temperature record used at the time showed approximately 0.3°C of cooling from around 1940 to 1970. But that record covered the region from 20°N to 90°N. It was weighted toward the Northern Hemisphere and was not interchangeable with a complete global temperature record. NASA’s Earth’s Temperature Tracker explains both the observed interval and the limitations of the data available at the time.

That distinction matters whenever a historical graph is used as evidence. A graph can accurately display a decline within its stated area and period while still failing to answer a broader question about global climate. Readers should check the geographical coverage, start and end dates, variable being measured and later revisions before accepting the interpretation attached to a chart. The same basic checks apply when using our checklist for evaluating a climate graph online.

Scientists were also studying two human influences that push temperature in opposite directions. Carbon dioxide reduces the rate at which Earth loses heat to space, producing warming. Some aerosols scatter incoming sunlight and affect clouds, producing an overall cooling influence. The balance depends on emissions, atmospheric chemistry, particle properties, location and time. Aerosols are short-lived compared with carbon dioxide and are distributed unevenly, which makes their climatic effects especially complex.

This was not merely a 1970s concern that later disappeared. The Intergovernmental Panel on Climate Change assesses that human-produced aerosols have exerted an overall cooling influence relative to 1850–1900, while also emphasizing their regional variation and interactions with clouds. The IPCC assessment of short-lived climate forcers describes the evidence and uncertainty behind that conclusion.

What the famous 1971 aerosol paper actually calculated

One paper often presented as an ice-age prediction was published in Science in 1971 by S. I. Rasool and Stephen H. Schneider. The authors calculated that a fourfold increase in the global background concentration of aerosols could reduce surface temperature by as much as 3.5 kelvin. If that condition persisted, they wrote, the resulting cooling could be sufficient to trigger an ice age. The same paper also discussed warming caused by additional carbon dioxide.

The words “could” and “if” carry much of the scientific meaning. The calculation tested the consequences of a large aerosol increase; it did not forecast that aerosol concentrations would actually quadruple. A conditional model result has the form “if X occurs, the model produces Y.” A forecast must also establish how likely X is, when it will occur and how other influences will evolve.

The paper’s assumptions and model can be criticized or improved, as happens throughout science. But rewriting a conditional experiment as “scientists predicted an ice age” changes the type of claim being made. It removes the assumed aerosol increase and presents a sensitivity calculation as a time-specific forecast.

Uncertainty was visible in the scientific assessments

Institutional assessments from the decade also contradict the idea of a settled cooling forecast. The U.S. National Research Council’s 1975 report, Understanding Climatic Change: A Program for Action, emphasized the need for better observations, analysis, modelling and theory. Its recommendations reflect a field with limited predictive capacity and major research needs, not an academy declaring that an ice age was about to begin. Readers can inspect the National Academies’ copy of the 1975 report directly.

Scientific uncertainty is not evidence that every proposed outcome had equal support. It means researchers were trying to quantify several influences with sparse observations and comparatively simple models. Some cooling scenarios were physically plausible under their assumptions. CO2-driven warming was also being studied, and researchers including Mikhail Budyko projected that increasing carbon dioxide would eventually overcome the mid-century cooling tendency.

By 1979, the National Research Council’s Carbon Dioxide and Climate assessment—usually called the Charney report—estimated equilibrium warming of roughly 1.5°C to 4.5°C for a doubling of atmospheric CO2. Its calculation section reported a range of 1.6 to 4.5 kelvin, with 2.4 kelvin considered likely within the assessment’s framework. That range would become an important benchmark in later climate assessments.

Why media coverage can create a different impression

The claim survives partly because magazine stories are easier to remember and circulate than technical papers. Newsweek published “The Cooling World” on April 28, 1975. Its author later said that the climate was warming and asked readers not to use the old article as evidence for the opposite conclusion. The article is real, but one prominent media story cannot establish the state of an entire research field.

Other supposed evidence is less reliable. A widely shared image resembling a 1977 TIME cover and carrying the headline “How to Survive the Coming Ice Age” is not an authentic TIME cover. Its fabrication does not erase genuine 1970s coverage of cooling. It simply means that this particular image cannot be used as historical evidence.

This is why source type matters. A peer-reviewed model experiment, a scientific assessment, a news report and a retrospective social-media image answer different questions. Treating them as interchangeable makes it possible to construct almost any historical narrative. A sound climate-claim fact-check starts by finding the original source and identifying what kind of document it is.

What happened to the mid-century cooling trend?

The observed trend changed. NASA describes the mid-century cooling interval as reversing around 1970. The IPCC now assesses with high confidence that global surface temperature increased faster from 1970 onward than in any other 50-year period in at least the previous 2,000 years.

That reversal is not, by itself, the entire case for human-caused warming. Modern attribution combines multiple lines of evidence: thermometer and ocean observations, satellite measurements, the known radiative effects of greenhouse gases, patterns of warming through the atmosphere and oceans, palaeoclimate records, and model experiments that compare natural and human influences. The IPCC’s synthesis explicitly draws on observations, physical process understanding, palaeoclimate evidence and climate modelling. Its Working Group I summary sets out how those lines of evidence support the conclusion that human influence has warmed the atmosphere, ocean and land.

This distinction prevents another common mistake. A short-term trend can motivate a research question, but it does not alone determine the long-term climate response. Scientists must identify the mechanisms producing the trend and test whether those mechanisms explain other observations. Mid-century cooling was not ignored; it became part of the evidence used to understand the combined effects of greenhouse gases, aerosols and natural variability.

A practical checklist for checking the claim

When someone presents a 1970s quotation, graph or cover as proof that scientists predicted an ice age, ask five questions:

1. Is the item authentic? Trace screenshots and quotations to the original publication rather than relying on a repost.
2. Is it a scientific paper, an assessment, a news report or an editorial? Each provides different evidence about scientific understanding.
3. Does the source make a forecast, or does it calculate what would happen under a stated condition? Preserve words such as “if,” “could” and “assuming.”
4. What period and geographical area does the graph cover? A Northern Hemisphere record or a few decades of data should not be silently relabelled as the entire global climate record.
5. What did the wider literature say at the time? One memorable paper or headline cannot demonstrate consensus; reviews and contemporary assessments provide the necessary context.

The claim confuses debate with consensus

The strongest historically accurate version of the story is straightforward: during the 1960s and 1970s, some scientists investigated whether aerosol pollution and observed mid-century cooling could lead to further cooling. A small number of papers anticipated cooling, while more anticipated warming. Scientific institutions emphasized substantial uncertainty and the need for better observations and models. By the end of the decade, a major assessment had quantified significant warming from doubled CO2.

That history does not show science mechanically switching from an ice-age consensus to a warming consensus. It shows researchers testing competing influences and revising conclusions as measurements, emissions trajectories and models improved. The practical lesson is not that old science was infallible. It is that a claim about what “scientists predicted” should be checked against the distribution of published research, the wording of original papers and the assessments available at the time—not against the most dramatic surviving headline.

References

1. THE MYTH OF THE 1970s GLOBAL COOLING SCIENTIFIC CONSENSUS
2. The myth of the 1970s global cooling scientific consensus – NERC Open Research Archive
3. Earth's Temperature Tracker – NASA Science
4. Chapter 6: Short-lived Climate Forcers | Climate Change 2021: The Physical Science Basis
5. Atmospheric carbon dioxide and aerosols: effects of large increases on global climate.
6. Read "Understanding Climatic Change: A Program for Action" at NAP.edu
7. On present‐day climatic changes – BUDYKO – 1977 – Tellus – Wiley Online Library
8. Carbon Dioxide and Climate: A Scientific Assessment
9. Stop Citing My 1975 'Global Cooling' Article | RealClearScience
10. 1977 "coming ice age" Time magazine cover is a fake – Science Feedback
11. IPCC AR6 Working Group 1: Summary for Policymakers | Climate Change 2021: The Physical Science Basis