In the late spring of 1893, in a small office at the Royal Greenwich Observatory in London, the assistant astronomer Edward Walter Maunder was working his way through the observatory’s two-century-old sunspot records as part of a routine archival project when he noticed something that, by his own later account, took him about four hours to fully believe.
Between approximately 1645 and 1715 — a span of seventy years — the historical sunspot count on the surface of the Sun had been almost exactly zero. Not “low.” Not “below average.” Almost exactly zero. The years for which the Greenwich records had usable data showed entire decades passing with fewer than ten observed sunspots — compared to the modern average of approximately a thousand observed sunspots per year. Maunder, sitting at his desk in 1893, was looking at the documentary record of a seventy-year period during which the most predictable feature of the Sun’s eleven-year cycle had simply not happened. The Sun, in the late seventeenth century, had been almost blank.
Maunder published the observation in 1894 in the popular astronomy magazine Knowledge. The reaction was, in his own description, “polite incuriosity.” The astronomical community of the 1890s did not believe him. The seventy-year gap was attributed by his peers to a defect in the historical records — possibly the result of bad telescopes, indifferent observers, or political disruption during the period (the English Civil War, the Thirty Years’ War, the Glorious Revolution). The idea that the Sun itself had actually been quiet for seventy years was rejected as more extreme than the underlying data warranted.
Maunder spent the rest of his career trying to get the result accepted. He never succeeded. He died in 1928. The seventy-year gap he had identified was largely forgotten by professional astronomy for the next forty-eight years.
How the gap came back
The rediscovery happened in 1976, in a single paper published in Science by the American solar physicist John A. Eddy of the High Altitude Observatory in Boulder, Colorado. Eddy had been working on the long-term variability of the Sun for several years and had become curious about whether the Maunder observation could be cross-checked against independent data. He had two new sources Maunder hadn’t had access to.
The first was the carbon-14 record preserved in tree rings. The amount of carbon-14 in the atmosphere is inversely related to solar activity — when the Sun is active, its magnetic field deflects more cosmic rays, which reduces carbon-14 production; when the Sun is quiet, more cosmic rays hit the atmosphere and produce more carbon-14. Tree rings record the atmospheric carbon-14 of each growth year. Eddy looked at the carbon-14 record for the period 1600-1800 and found a sharp spike between approximately 1645 and 1715 — exactly when Maunder said the sunspots had been absent. The spike was independent of any direct sunspot observation. It was a different kind of evidence pointing to the same conclusion.
The second source was the aurora borealis records. The aurora is also a product of solar activity — solar particles striking the Earth’s upper atmosphere produce the visible aurora at high latitudes. When the Sun is active, auroras are frequent and reach lower latitudes; when the Sun is quiet, they are rare and confined to the polar regions. Eddy examined European aurora records from 1600-1800 — preserved in observatory archives, in monastic chronicles, and in the records of the Russian and Norwegian fishing fleets — and found a sharp drop in aurora frequency that matched the sunspot gap almost exactly. The Greenwich observatory itself recorded no aurora visible from London between 1645 and 1715, despite extensive monitoring.
The three independent records — sunspots, tree-ring carbon-14, and aurora frequency — agreed precisely. Eddy concluded that the Maunder observation was correct: the Sun had been almost completely inactive for seventy years.
The 1976 paper was, by general consensus, the moment that long-term solar variability became an accepted research area in astrophysics. The Maunder Minimum (Eddy coined the name in the paper, in tribute) became one of the foundational observations of the modern field. Eddy’s paper also identified earlier minima — the Spörer Minimum (1450-1540), the Wolf Minimum (1280-1350) — and noted that these correlated with cold periods in European climate records. The link between solar variability and terrestrial climate, which had been on the speculative fringes of climatology, became a serious subject of study.
What it did to the weather
The Maunder Minimum coincided almost exactly with the coldest portion of the Little Ice Age in northern Europe. The seventeenth-century winters that are now famously cold in European history — the 1684 winter in which the Thames froze for two months, the 1709 Grand Hiver when the Seine froze and most of France’s olive trees died, the recorded winter scenes by Dutch painters of skaters on frozen canals that no longer freeze — all happened within the Maunder period.
The direct climate effect of reduced solar activity is modest. The total solar irradiance during the Maunder Minimum has been estimated, on the basis of subsequent satellite calibrations against the 1976-2020 minor minima, to have been approximately 0.2% lower than the modern average — a small change that, in modeling, produces an average global cooling of perhaps 0.1°C. The actual cold of the Maunder period in northern Europe — winter temperatures roughly 1.0 to 1.5°C below modern averages — was substantially larger than the direct solar effect could explain.
The current best modeling, as set out in the work of Mike Lockwood and colleagues at the University of Reading in the 2010s, is that the Maunder cooling was the result of amplification: the small direct solar cooling triggered larger climate feedbacks, particularly involving the North Atlantic ocean circulation and sea-ice extent. The Sun set the conditions; the ocean and ice did most of the cooling.
This is also the modern understanding of how solar variability affects climate generally. The direct effect is small; the indirect effects can be substantial. Modern modeling suggests that even a substantial future solar minimum would produce only modest additional cooling — perhaps 0.1 to 0.3°C — which is small compared to the warming from continuing anthropogenic greenhouse-gas emissions. The Maunder Minimum, in other words, was a real and large cooling event, but it does not provide a path back from modern warming.
Where Maunder’s notebooks are
The original Greenwich Observatory sunspot records that E. Walter Maunder examined in 1893 survive in the Observatory’s archive, now held at the Cambridge University Library. They are bound notebooks containing daily sunspot drawings made by successive Greenwich observers from approximately 1675 through the late nineteenth century. The seventeenth-century portion is the work primarily of John Flamsteed, the first Astronomer Royal, who recorded almost no sunspots between 1675 (when he began routine observations) and the end of his term in 1719.
Maunder’s own working notes — including the original 1893 marginalia where he first identified the gap — are in a separate small archive at the Royal Astronomical Society in Burlington House, London. The notes can be examined by appointment. The marginal note next to the entry for the year 1671 reads, in Maunder’s hand: Not one sunspot. Cannot be right. Must be records.
He was right. The records were right too.