Science Briefs

Glaciers, Old Masters, and Galileo: The Puzzle of the Chilly 17th Century

Figure 1: Painting of skaters on a frozen canal.
Figure 1: Sports on a Frozen River, by Aert van der Neer (courtesy The Metropolitan Museum of Art).

During the heart of the period known as the "Little Ice Age" during the 17th Century, temperatures were extremely low in much of Europe and what would become the eastern United States. Dutch masterpieces from that time by Brueghel, van der Neer and others (Figure 1) show people skating on canals and rivers that almost never freeze today. Ice was so prevalent in northern seas that Inuit were seen fishing as far south as Scotland. Glaciers descended from the Alps, destroying outlying farms and threatening to crush whole villages. In North America, native tribes banded together to form the League of the Iroquois in the face of declining food supplies and other natural hardships during these cold years.

It has long been speculated that the drop in temperatures was due to a dimmer Sun. After Galileo's popularization of the telescope in 1609, he and several other early astronomers soon observed and studied dark spots on the Sun. These "sunspots" cycled over a period of about 11 years, as they do today (Figure 2), but after 1645, prominent sunspots almost totally vanished. They reappeared around 1715, and the sunspot cycle has been present ever since. The decades with almost no sunspots is now called the Maunder Minimum. Modern measurements have confirmed the early assumption that the number of sunspots is related to the total brightness of the Sun.

Fig 2a: Animation of Galileo's sunspot observations. 310 kB MPG Fig 2b: Animation of modern sunspot observations. 6.3 MB MPG
Figure 2: Sample images from animations of sunspor observations. At left, Galileo's sunspot drawings from 1611; click for 310-kB MPG movie (courtesy Albert van Helden, Galileo Project, Rice University). At right, modern satellite observations; click for 6.3-MB MPG movie (courtesy NASA/ESA).

The reduction in solar output during the Maunder Minimum was about one-quarter of one percent, though it is difficult to determine this value exactly. Although this seems a very small change, the output of the Sun is so large that this can still have a sizeable impact. However, it is not enough to plunge the whole Earth into "Little Ice Age" type conditions. Whether the dimmer Sun can indeed explain the extreme cold during the 17th century has therefore been a puzzle.

Based on climate modeling, we have proposed a solution to the apparent paradox of extreme cold with only a marginally dimmer Sun. In our simulations, we find that the reduced brightness of the Sun during the Maunder Minimum causes global average surface temperature changes of only a few tenths of a degree, in line with the small change in solar output. However, regional cooling over Europe and North America is 5-10 times larger due to a shift in atmospheric winds.

We compared the climate model results to surface temperatures during the Little Ice Age. Since little thermometer data exists from that time, we relied upon indirect temperature information from tree rings, ice cores, corals, and historical records. Global average temperature changes are small in both the climate model and the data. Both also show that surface temperature changes associated with solar output changes exhibit alternating warm oceans and cold continents at Northern Hemisphere mid-latitudes (Figure 3). In the model, these occur primarily through a slowdown in the speed of westerly winds at the Earth's surface. Greater heating by the Sun in the tropics relative to high latitudes causes an equator-to-pole flow of air, which is turned towards the east by the Earth's rotation.

Figure 3: N. Hemisphere temperature maps. See caption.
Figure 3: Annual average surface temperature change (C) due to solar irradiance change between the Maunder Minimum (late 17th century) and a century later, when solar output had returned to relatively large values, in the climate model (top) and in the historical temperature reconstructions (bottom). Click for larger version.

So a reduction in the amount of sunlight reaching the planet leads to a weaker equator-to-pole heating difference, and therefore slower winds. The effect on surface temperatures is particularly large in winter. Because the oceans are relatively warm during the winter due to their large heat storage, the diminished flow creates a cold-land/warm-ocean pattern (Figure 3) by reducing the transport of warm oceanic air to the continents, and vice-versa.

Changes in this wind flow have only a small impact on global temperatures as the warm and cold regions average out, but they have large regional effects. They also increase the frequency of extreme events, so that the modeled reduction in winds would lead to many more extremely cold days over Europe and eastern North America (which may stand out in the historical record).

These results begin to reconcile the longstanding dilemma of how the change in solar output could have been very small and yet have led to much colder temperatures in Europe and eastern North America, the areas from which the historical evidence for the "Little Ice Age" originates.


Shindell, D.T., G.A. Schmidt, M.E. Mann, D. Rind, and A. Waple 2001. Solar forcing of regional climate change during the Maunder Minimum. Science 294, 2149-2152.