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.
At the beginning of the twenty-first century, however, Croll’s astronomical theory is again alive and well. It has been refined and extended, but the basic premise is the same. Strangely, however, Croll himself is rarely mentioned, even in many modern textbooks. The orbital cycles, evidence for which can be seen very clearly in the changing sediments of deep sea cores, are usually referred to as “Milankovitch Cycles,” named for the Serbian mathematician Milutin Milankovitch, who, to be fair, put the theory on a modern footing and is responsible for its renewed widespread acceptance. We shall see how he did that later in this book. Nonetheless, one has the feeling that Croll is a bit of a forgotten hero in the ice age story. Very occasionally, the astronomical theory of ice ages is referred to as the “Croll-Milankovitch” theory. Milankovitch, in his own writings, fully recognized Croll’s contributions. Perhaps Croll will eventually get the recognition due to him from others as well.
CHAPTER SIX
Defrosting Earth
Joseph Adhémar’s apocalyptic vision of a melting polar ice cap, a shifting center of gravity for the Earth, and devastating floods as ocean water rushed from one hemisphere to another is far from an accurate picture of what happens at the end of a glacial period. But at their maximum, the ice caps did hold huge volumes of water, and accumulating evidence shows that there were glacial floods as they melted, some of them catastrophic. In fact, some were so catastrophic that when the claim was first made that flowing water had produced the devastation they left in their wake, the idea was dismissed as preposterous. This reception was eerily reminiscent of the initial reaction to Agassiz’s ice age theory. Catastrophic flooding associated with melting glaciers is now universally accepted, however, and in recent years, the topic has attracted renewed interest. Not only did the gigantic floods wreak havoc with the landscape, but they may also have been responsible for drastic shifts in climate.
The tale of gigantic glacier-related floods begins in the Pacific Northwest of the United States. In the state of Washington, there is a tract of land that early settlers called the scablands. It is harsh, dry land that to them was reminiscent of a partially healed wound, the soil ripped away and the countryside gouged and scarred by some unknown force. The unknown force, it turns out, was water from melting glaciers, water released in floods that were orders of magnitude larger than any known in human experience, so large that they completely transformed the landscape over which they flowed. Again reminiscent of the ice age theory, the idea that there was a connection between Pleistocene glaciation and the scablands was championed largely by one man, a University of Chicago geologist named J. Harlan Bretz. Bretz was a field man and an empiricist. Like many geologists, he considered his field evidence in the context of “multiple working hypotheses.” The hypothesis that fit his data best, he concluded, was that the bizarre landscape of the scablands had been produced by one or more huge floods associated with the melting of ice age glaciers. Some of the features he studied could be explained by less dramatic processes, he admitted, but the only way to understand the totality of the scablands landscape was to invoke a flood of such magnitude that it was unique in the geologic record. In a paper written in 1928—five years after his first major published work on the scablands, and at a time when his ideas were being vehemently criticized by prominent geologists—Bretz wrote: “The region is unique: let the observer take the wings of the morning to the uttermost parts of the earth: he will nowhere find its likeness.”
Bretz was born in Michigan in 1882. For a while he worked as a schoolteacher, first in Michigan and then in Seattle, Washington. It was there that he began his systematic studies of glacial geology, spending weekends and summer holidays tramping around the Puget Sound area and keeping detailed notes of his field observations. Eventually he decided to return to university to study for an advanced degree. Using the field notes he had accumulated in Seattle as the basis for a thesis, he graduated with a Ph.D. summa cum laude from the University of Chicago in 1913. Not long afterwards, he accepted a faculty position at his alma mater, where he was to spend the rest of his career. But living in the Midwest didn’t divert his interest from the problems of glacial geology in the state of Washington. He quickly organized a summer field course for students, whom he took to the gorge of the Columbia River in southern Washington, where they studied, among other things, erratic boulders that they found distributed to elevations several hundred meters above the highest recorded water levels of the river. But unlike the erratic boulders of the Alps that were so important for Agassiz’s theory, those along the Columbia River gorge could not have been carried by flowing ice—geologists had already established from moraines and other evidence that the Pleistocene glaciers had never reached that far south. The only plausible explanation was that the boulders had been carried to the gorge embedded in large floating rafts of ice that had broken off from glaciers hundreds of kilometers to the north. If that were the case, the distribution of the erratics would require very deep water, much, much deeper than the present day Columbia River. To explain the ice-borne erratics, some geologists concluded that this part of Washington had been submerged under the waters of the Pacific when they were deposited. But Bretz could find no evidence of ancient beaches or marine fossils, so he concluded that the icebergs had floated in fresh water. The first seed of the idea that there had been an unimaginably large flood as the Pleistocene glaciers melted away was planted in his brain.
Year after year for more than a decade Bretz took his University of Chicago students back to Washington for summer fieldwork. Especially in the early years, much of their travel was on foot. Later, Bretz had a car, which meant more ground could be covered. They were a motley crew: in addition to his students, Bretz was usually accompanied by his wife and children and the family dog. But motley crew or not, Bretz and his students gradually built up a detailed and intimate knowledge of the glacial geology of the area. After a few years concentrating on the Columbia Gorge in southern Washington, Bretz decided to move north to the scablands. Ever since his days as a schoolteacher in Seattle, he had been fascinated by the strange morphology of the landscape revealed on the topographic maps of the area—especially the giant “Potholes” waterfall, which sits several hundred meters above the present-day water level of the Columbia River. What Bretz and his students encountered in the scablands was, to Bretz at least, mind-boggling. To the untrained eye, it was a chaotic landscape of channels, basins, and hills. But when he had put all the pieces together, Bretz understood that he was looking at the work of a volume of flowing water so immense that it had eaten its way through hundreds of meters of loess, cut channels in solid basalt, and thrown up gravel bars so large that they made those of even the largest present-day rivers look positively Lilliputian.
Bretz added the adjective “channeled” to scablands, and ever since this vast area of eastern Washington has been known as the Channeled Scablands. He described it as a roughly rectangular block, bordered on the north and west by the deep canyons of the Columbia River, and on the south by the Snake. These two rivers merge in southern Washington and flow together westward to the Pacific. The Channeled Scablands are underlain by the dense, hard rock of the Columbia River Basalts, outpourings of lava that erupted in a flood of their own about sixteen million years ago and covered the preexisting landscape with layer upon layer of basalt flows. The Columbia River skirts along the northern and western edges of the basalt, but the floods that Bretz chronicled didn’t f
ollow the preexisting river course. Instead, they overflowed the banks of the Columbia and swept right across the basalt plain, scouring deep channels and potholes where none had been before.
Words and numbers can quantify the features that Bretz observed, but it is hard to grasp the magnitude of the floods that occurred without comparison to things that are more familiar. The floods swept across an area of the Columbia Plateau that is estimated to be at least 7,000 square kilometers. The peak water levels probably lasted no longer than days, possibly only hours. The amounts of water were prodigious—the best estimates suggest that more than ten million cubic meters of water per second coursed across the scablands. That is about sixty-five times the average discharge of the Amazon River today, and roughly ten times the average discharge of all the world’s rivers to the oceans. And this deluge was both brief and confined to a limited area in eastern Washington. It’s no wonder that the effects of the scablands floods baffled observers—no one had seen anything like them before.
Bretz must have arrived gradually at his conclusion that a huge glacial flood had created the Channeled Scablands. Certainly, the erratic boulders were an early clue, but it was the bizarre topography of the Scablands themselves that drove the message home. In his first published paper on the subject, which appeared in the Bulletin of the Geological Society of America and was innocuously titled “Glacial Drainage on the Columbia Plateau,” Bretz described the scabland features, but he was cautious about interpreting them. He notes the fact that the Columbia River swings around the northern edge of the basalt plateau rather than cutting directly across it. He describes the geological history—the evidence that during glacial times, the Pleistocene ice sheets reached the northern border of the basalt plain but advanced no farther. He explains that during the ice age, the entire plateau was mantled with a covering of wind-blown loess about seventy meters thick, and that sometimes ash from volcanic eruptions in the Cascades Range to the west is mixed in with the loess. Then he focuses on the details of the scablands—the myriad intertwining channels that cut down through the loess to the basalt below, and, in places, extend far into the basalt itself; the deep dry canyons or coulees, some with walls that are three hundred meters high; the gigantic gravel bars that are made up nearly entirely of pebbles of the local Columbia River Basalt. He says that all of this must be the result of erosion by water, but he doesn’t yet use superlatives, although he does refer to a glacial flood. He suggests that the water must have come from melting Pleistocene glaciers to the north, and that the ice must have dammed up the northern reaches of the Columbia River, because that would have forced the water to drain southward over the plateau rather than following the course of the river. Reading between the lines of his clear and measured prose, it is evident that Bretz already realized that the volumes of floodwater must have been enormous.
Over the several years following that 1923 paper, Bretz became increasingly confident in espousing and defending his theory that the Channeled Scablands had been produced by catastrophic flooding. He continued to build up evidence from his field studies, and nearly every new observation seemed to bolster his theory. But there were many critics and very few proponents of his ideas. Geology by catastrophe was anathema to most geologists of the early twentieth century. The textbooks and conventional wisdom dictated that geological processes took place gradually, and that the carving of river valleys by streams was something that happened only over very long periods of time. In spite of Agassiz and his catastrophist ideas about the ice age, the gradualist view of nature had been in vogue for about a century, and for the most part, it had served geology well. Most geological processes are slow and gradual. Mountain ranges, the Grand Canyon, sand on a beach—these are all things that form over times that are very long from a human standpoint. Even the waxing and waning of the Pleistocene glaciers was understood by Bretz’s time to be a gradual process, not the flash-freeze that Agassiz had envisioned, freezing woolly mammoths in their tracks. The Scabland floods, on the other hand, had to have been catastrophic. In Bretz’s view, there was no alternative, and he saw no real contradiction in the possibility that the rates of geologic processes could occasionally change drastically. He realized that this would raise strong objections from those conditioned to thinking about erosion and deposition—the two processes that had created the unique landscape of the Scablands—as gradual processes. Once again reminiscent of Agassiz before him (although a bit more modestly), Bretz stated this explicitly when he wrote about his theory. In 1928, writing again in the Bulletin of the Geological Society of America, he observes that strong opinions can distort both sides of an argument: “Ideas without precedent are generally looked on with disfavor and men are shocked if their conceptions of an orderly world are challenged. A hypothesis earnestly defended begets emotional reaction which may cloud the protagonists’ view, but if such hypotheses outrage prevailing modes of thought the view of antagonists may also become fogged.”
Bretz rested his case on the field evidence. He would urge his critics—many of whom had no firsthand knowledge of the features he had described—to travel to Washington and view the Scablands themselves. Undoubtedly, he would have agreed with something Agassiz was reputed to have told his students: “Study nature, not books.”
Bretz seems to have been a systematic and well-organized person. He was fond of lists. His papers often contain a succinct enumeration of all of the features he observed in the field. A partial inventory of the observations that led him to conclude that the Scablands had resulted from a catastrophic flood includes the following, paraphrased from his own lists:
• All of the areas of Scabland on the Columbia plain are connected to one another. The channels all slope downward from the northeast toward the Snake and Columbia Rivers to the south and west, and they are “anastomosing” or braided, unlike most river systems in which tributaries flow into a main stream in a dendritic pattern.
• There are just ten “openings” to the Scablands channels from the north, and nine exits where they drain into the Snake and Columbia Rivers.
• The loess, which still blankets other parts of the Columbia River Basalt plain, is almost completely stripped away in the Scablands tracts, and where it remains it is in the form of elongate, teardrop-shaped hills oriented parallel to the local Scabland channels.
• Although the basalt plain extends beyond the Columbia River in the west, and the Snake River to the south, no Scablands features exist there.
Even this partial list of Bretz’s evidence for flooding is persuasive. Although on a different scale, the interfingering pattern of the channels in the Scablands—anastomosing as Bretz called it—is characteristic of most rivers that experience periodic flooding; they are aptly referred to by geologists as braided streams. The teardrop-shaped hills of loess in the Scablands have steep sides that show signs of erosion by water, and their overall pattern suggests a flotilla of ships, all with their prows pointing upstream. The tops of these hills have soils developed on them, and the surfaces of some of the larger ones have their own small but normal drainage systems, completely unlike the braided channels that separate them. Evidently, the floodwaters did not reach the tops of these hills. The fact that all of the channels empty into the major rivers to the south and west indicates that the floodwaters were eventually carried out to the Pacific by the Snake and Columbia Rivers.
Bretz didn’t use hyperbole, he simply presented his observations and sought the best explanation for them. The scale of the Scabland features, he believed, required voluminous, short-duration flooding. For the most part, no one questioned his observations, but his conclusions were strongly opposed. In January 1927, he was invited to present his views to the Geological Society of Washington, D.C. In the audience were many of the most distinguished geologists of the day, and they had choreographed a series of negative responses to what they anticipated Bretz would say. Once again, he laid out his arguments. By this time, his language had become more confident and forceful. Ag
ain he used bulletlike lists:
• Canyons of the Scablands. Largely channels of huge rivers.
• Rock basins in the channels. Thousands of them. . . . Lengths as great as eight miles, depths as great as 200 feet. . . . Formed by large vigorous streams plucking the . . . basalt.
• Cataracts. Hundreds of extinct waterfalls . . . several two to three miles wide.
• Trenched divides [here he is referring to elevated tracts of the landscape that separated different drainage systems but were “trenched”—cut across—by the floodwaters]. Several remarkable cases where . . . gashes 200 to 400 feet deep [cut] across a divide. . . . Water must have been 100 to 300 feet deep above the preglacial valley bottoms on the north to have crossed.
• 100 to 200 feet of loess removed over large areas.
• Contemporaneous occupation [by water] of all Scabland routes seems indicated. . . . Anastomosis due to the huge volume of glacial water and [its] abrupt introduction.
Although Bretz’s arguments were impressive, there was one major weakness that many of his critics seized upon. Bretz was very much aware of the difficulty: he admitted that he had no reasonable explanation for the very sudden production of the huge amount of glacial melt-water required by his hypothesis. It was possible, he said, that a volcanic eruption had occurred under the ice to the north of the Scablands, resulting in massive melting, but there was no known independent evidence for such an event.
Frozen Earth: The Once and Future Story of Ice Ages Page 12