Frozen Earth: The Once and Future Story of Ice Ages (15 page)

Croll was influenced by Adhémar’s book, particularly by the idea that the combined effects of eccentricity and the wobble of the Earth’s axis might be important for glaciation.
However, he showed that there were errors in Adhémar’s calculations, which, in his view, invalidated some of the conclusions.
Furthermore, “the somewhat extravagant notions which Adhémar has advanced,” as he put it—namely, the collapsing ice pedestal and shifting center of gravity—were just a bit too fantastical for the serious Scot.

If others had already written about possible astronomical causes for the ice age, what then was Croll’s contribution to this scientific problem, and why was it so important?
The key, I think, was his determination to understand the process from first principles.
His very first paper on the subject, which appeared in the August 1864 issue of
Philosophical Magazine and Journal of Science,
was masterful.
He enumerated and evaluated each of the various theories that had been proposed to explain the ice age, and then summed up as follows: “Another objection which
we have to all these hypotheses which have come under our consideration is, that every one of them is irreconcileable
[sic]
with the idea of a regular succession of colder and warmer cycles.
The recurrence of colder and warmer periods evidently points to some great, fixed, and continuously operating cosmical law.”

The geological evidence for alternating glacial and interglacial episodes was by this time quite clear, and Croll realized that it could not be ignored.
“The true cosmical cause must be sought for in the relations of our Earth to the sun,” he went on to suggest.
The fundamental principle, he recognized, is that solar energy is the external force that regulates the Earth’s climate.
For more than a decade, “the relation of our Earth to the sun” was to be the focus of Croll’s research.

Croll found that slight variations in the Earth’s orbit around the sun could not themselves be responsible for initiating or ending an ice age, because the resulting fluctuations in solar energy received at the Earth’s surface were just too small to cause significant changes in average temperature.
But unlike others who had reached a similar conclusion, he did not immediately dismiss the idea.
The cyclical nature of both the orbital variations and ice ages convinced him that there must be a connection, so he turned to the problem of how the effects of small orbital changes might be amplified.
He began by examining how heat is distributed on the Earth, and concluded that ocean currents are extremely important in determining the heat balance between the Northern and Southern Hemispheres—important enough that changes in ocean currents could initiate or end glaciation in the Northern Hemisphere.
The orbital variations, he argued, while too small by themselves to cause large temperature changes, could affect ocean currents and lead to glacial cycles.
Secondly, having identified a mechanism, he carried out the painstaking calculations referred to earlier, computing how the shape of the Earth’s orbit has changed over a period extending from three million years ago to a million years into the future.
His graphs showed regularly alternating peaks and valleys, with many of the large peaks recurring at roughly 100,000-year intervals (figure 11).
The heights of
the peaks varied considerably.
Croll identified those times in his record when the orbit was most elliptical as the most likely glacial periods.

Figure 11.
Part of James Croll’s graph of the changes in eccentricity of the Earth’s orbit around the sun.
Croll believed that ice ages could only occur when the eccentricity was high, with values approaching the upper solid line on his graph.
On this basis, he concluded that the most recent ice age had ended about 80,000 years ago.
Graph based on data from Croll,
Climate and Time in Their Geological Relations: A Theory of Secular Changes of the Earth’s Climate
(London: Daldy, Isbister, 1875).

In what was really an extension of Adhémar’s ideas, Croll reasoned that it was the interaction between the Earth’s wobble and the changing eccentricity of its orbit that was most important.
He asked himself what would happen if both the eccentricity was at a maximum
and,
because of the precession of the equinoxes, the Northern Hemisphere winter occurred when the Earth was farthest from the sun in its orbit.
He concluded that the winters would be very much colder than at present.
He also concluded that the cold temperatures would extend farther south, and that snow would fall and remain at more southerly latitudes than occurs today.
This, in turn, would enhance the cooling, because snow reflects more of the sun’s energy than bare ground.
He also proposed that the greater temperature contrast between the equatorial and polar regions would increase the winds, which would affect ocean currents in such a way that the amount of heat carried northward by the Gulf Stream would diminish.
The increase in wind intensity would also, he reasoned, carry more moisture from tropical regions toward northern latitudes, where it would precipitate and add to the accumulating blanket of snow.
All of these factors—heat carried by ocean currents, the reflecting power of snow cover, and the availability of atmospheric
moisture for precipitation—were first clearly demonstrated by Croll and are still recognized as important factors for the buildup of continental ice sheets.

Finally, Croll pointed out an obvious but often overlooked consequence of his theory: if he was correct, it could be used as a geological chronometer.
When Croll did his work on ice ages, radioactivity had not yet been discovered, and geologists had no way to measure time accurately.
The age of the Earth was unknown, although there were some wildly varying estimates based on such things as how long it would take an originally molten Earth’s crust to cool to its current temperature.
But the astronomical variations could be calculated quite accurately as a function of time.
If changes in climate were really tied to the changes in the Earth’s orbit, the geological effects—the glacial drift, the interglacial fossils and soils—could be dated simply by using Croll’s graphs of orbital variations.

Although there are details of Croll’s theory that were later found to be incorrect, it was nevertheless a remarkable achievement.
This self-taught intellectual was the first to recognize the multiple interconnections in the Earth’s climate system, something that is very much at the forefront of modern research.
He anchored his theory in the astronomical variations, but realized that they are probably a trigger, rather than a cause, for glacial-interglacial cycles.
In the most general sense, that is essentially the state of our understanding today.

Croll’s papers on heat and electricity had brought him some recognition, but his astronomical theory for ice ages suddenly made him a very important figure in the world of nineteenth-century science.
Fortuitous timing may have helped his rise to prominence, because there was intense interest in glaciation and its effects.
The debate about the reality of an ice age had moved on; now the focus was on mapping the distribution of the glacial deposits, figuring out the sequence of glacial cycles, and, above all, determining the reason the climate had shifted in the first place.
Croll was at the center of a hot field of enquiry, and the most eminent scientists of the day sat up and took notice.

Among those impressed by Croll’s paper in
Philosophical Magazine
was Archibald Geikie, mentioned earlier, who had written extensively on glaciation in Scotland and who had just been appointed director of the recently reorganized Scottish Geological Survey.
Geikie offered Croll a job.
But Croll wrote to Geikie that he “did not see [his] way clear to accept the proposal.”
He was, he said, “somewhat up in years” (he was all of forty-three in 1864), but perhaps more important, he was not in the best of health, and he had no background in geological matters.
He wasn’t sure he had the necessary qualifications for the post.

Geikie was persistent, however, and he eventually persuaded Croll to move to the Edinburgh headquarters of the survey, not as a senior scientist, but as a “resident surveyor and clerk.”
Croll first had to take the Civil Service examinations—parts of which he failed—before his appointment became official, but in September 1867, he took up his new position.
It was to be the last place of employment in his wandering career; he retired from the Scottish Geological Survey thirteen years later.
During his time with the survey, he was showered with honors: an LL.D.
degree from the University of St.
Andrews, election as a Fellow of the Royal Society of London, and election as an Honorary Member of the New York Academy of Sciences.
He was also awarded a number of scientific prizes.
His official work was not very onerous, and he continued his “private work” in the evenings, writing papers and eventually a landmark book that brought together his ideas on climate change:
Climate and Time in Their Geological Relations: A Theory of Secular Changes of the Earth’s Climate,
published in 1875.

Croll’s influence on thinking about glaciation was at its peak.
Many leading geologists took up his theory, recognizing that it was the best way to explain the geologically rapid alternation of glacial and interglacial periods that was implied by the field evidence.
But as time went on and more evidence about the ice age accumulated, doubts began to surface.
There were two main problems raised by the critics.
The first was that Croll’s theory, like Adhémar’s, predicted that the hemispheres would be glaciated alternately every 11,500 years, half the length of the
Earth’s wobble cycle.
Secondly, Croll proposed that glaciation would only occur when the eccentricity of the Earth’s orbit is at its maximum.
Under these conditions, ice would form in the hemisphere that was tilted away from the sun—the winter hemisphere—when the Earth was also farthest from the sun in its orbit.
Because the eccentricity cycle is much longer than the wobble cycle, eccentricity could remain large enough for each hemisphere to experience a series of glacial and interglacial cycles.
As the eccentricity slowly decreased and the orbit became more nearly circular, the freeze-thaw cycle of the ice age would end, and the entire planet would have a warmer and much more uniform climate.
Croll’s calculations showed that the most recent period of high eccentricity was almost exactly 100,000 years ago, and that since then the eccentricity has decreased fairly rapidly to its present value.
On this basis, he suggested that the last great glaciers must have retreated about 80,000 years ago.

Testing predictions is a key to scientific progress.
Croll’s theory could be tested if it were possible to date the deposits left by the ice age glaciers.
The ages of glacial drift in the Northern and Southern Hemispheres should differ by 11,500 years, and if Croll was right about eccentricity, the youngest glacial deposits should be about 80,000 years old.
Although there were no very accurate ways to measure geological time in Croll’s day, there were some clever ideas.
One was to use fossils.
Species evolve over time, so that the assemblage of fossils in a deposit is generally distinctive for a given time.
If the fossil assemblages in sediments at different places on Earth are similar, it generally means that they are about the same age.
Fossils are not very abundant in glacial deposits, but sometimes, by combining fossil evidence and the principle of superposition, it is possible to bracket, or at least compare, the ages of several generations of glacial drift.
When this approach was used for deposits in both the Northern and Southern Hemispheres, little evidence was found for different ages.
In conflict with Croll’s theory, there seemed to be no firm case for alternating glaciation in the two hemispheres.
But the observation that really caused support for
Croll’s theory to waver came from studies of erosion at Niagara Falls.
Below the Falls is a steep-walled gorge, which has been formed by the waterfall’s erosion of the rocks over which it flows.
By direct observation, American geologists had estimated that the Niagara gorge is being eroded back at the rapid rate of almost a meter per year.
This provided a way to estimate the age of the gorge, and, because the Niagara River flows across a deposit of glacial drift below the gorge, it also allowed a minimum age to be calculated for the glacial episode associated with the drift.
The time estimated in this way was 10,000 years, clearly at odds with Croll’s estimate that the last glacial episode had ended 80,000 years ago.
Even though there were large uncertainties in the Niagara Falls calculation, and even though the age was later revised upward—to as much as 30,000 years—the discrepancy sowed seeds of doubt about the validity of Croll’s astronomical theory of ice ages.

Croll still had many supporters, but they had to begin to make ad hoc arguments—perhaps, for example, local effects played a part, and thus the glacial records differed in Europe and North America.
But it was a losing battle.
In spite of the appeal of the astronomical cycles for explaining the alternation of warm and cold intervals, the predicted timing seemed to be wrong, and there was no conclusive evidence for alternation of hemispheric glaciation.
Gradually, Croll’s theory fell out of favor.
By the end of the nineteenth century, the astronomical theory was all but dead.