Letters from Grand Canyon—
Reversal: Establishing a Course

We have been following events that have occurred over hundreds of years in the area that is now the Colorado Plateau (traversed by the Colorado River). These events are reasonably clear in their overall architecture, but the fine details escape resolution, much as the far wall of the Grand Canyon, hazy in the shimmer of late-afternoon light, reveals the grand entablature of major cliffs and the contrast of dark forest and lighter bare rock, but not individual ledges and trees. We accept this soft focus of understanding because time and space wear down detail, as they should. But now we have arrived at last at the Cenozoic, the youngest era, the age of mammals, our age, a time when we might expect to be able to lean back with a contented sigh and say: “Now at last, we can see, know, understand everything, now we can finally relax.” But this is perhaps the most puzzling, the most obscure, the least well known part of the whole story.
The problem is this—as was discussed in the previous Letter (Genesis), Mother Earth had quite a sneeze at the end of the Mesozoic Era: tectonic plates crashed into each other, mountain ranges were thrown up, interior seas sloshed in and then back out, and the landscape was changed profoundly. When the dust settled, what is now the Colorado Plateau had acquired the general shape of a huge saucer, rimmed by mountains and sporting large lakes in the low-lying center. Not surprisingly, rivers, which have a marked preference for going downhill, took the sensible course of generally flowing radially inward from the mountainous rim to the lakes at the center. In the case of northern Arizona and adjacent southern Utah, this means that they flowed North, which is the exact opposite to the present course (generally South to Southwest). How, and when, precisely, did this reversal take place? What might have caused it? We don’t know (so much for the Apollonian clarity of vision). With luck, someday we will find a clue that will complete the puzzle, or put the information we have into a new framework that will enable us to say “of course!”
Although the broad outlines of the landscape of today were already in place shortly after the “sneeze,” the details were not, and it is these details which give geologists trouble. Our understanding is made more difficult because of the scarcity of deposits and landscape features which we can attribute to the time of drainage reversal. Without them, there is little to work with, and we can only make a proposal as to what occurred, and when.
The building of the proposal begins with the colorful geologist Charles B. Hunt. Hunt spent decades in parts of the Colorado Plateau that are today infested by motorhomes and mechanical contraptions of every kind, but in the ’30s, ’40s and ’50s were as empty and lonely as Tierra del Fuego. He got to know the plateau like the inside of his pockets, and was especially intrigued by the rounded river gravel, abandoned channels, and isolated remnants of valleys (such as Unaweep Canyon across the Uncompahgre Plateau of southwest Colorado) which he kept finding in places where no rivers are found today. With time, dedication, and much patience, Hunt assembled the various pieces into a commendable picture, which he published in 1969 (Geologic History of the Colorado River in U.S. Geological Survey Professional Paper 669, The Colorado River Region and John Wesley Powell.)
What Hunt found is that by the Miocene time (roughly 25 to five million years ago), the Colorado Plateau had a river network that could be considered ancestral Colorado River, and so on. The courses of the ancestral rivers departed considerably from the modern ones in detail, but the ancestral rivers flowed in the same direction as the rivers of today, south to southwest. This means that the reversal of flow probably took place somewhere in the interval between about 40 million years ago, when rivers still flowed into the central lakes, and the Miocene.
How did it happen? Hunt proposed that the southern course was established when the lakes overfilled and spilled through a low area to the south. This is not very realistic for a variety of reasons. In particular, the western and southern rim of the saucer consisted of the formidable mountain range, the great rim, produced by the “sneeze” that took place at the end of the Mesozoic and the beginning of the Tertiary. This would not be a probable place for the lowest spot of the entire rim. The lowest spot would most likely have been to the east, bordering the High Plains and producing eastward flowing rivers such as the Platte and Arkansas. So, we are left with two workable possibilities for reversing the course of the streams of the plateau:
a) Tilting of the Colorado Plateau
b) Eroding part of the rim to below the level of the lake.
Tilting seems like a promising way to lower the rim relative to the center and to reverse drainage. Even a very small tilt makes for very large differences in elevation over the horizontal distances that are in question here: a tilt of only one degree makes for a difference in elevation of nearly four and a half miles over a horizontal distance of 250 miles (roughly the distance between the lakes in Utah and the rim of the plateau). Thus, tilting the plateau to the south by one degree would lower the southern rim by four to five miles (or raise the center by a like amount), ample to achieve the reversal (Figure 1). But, there is a problem. The sediments which were deposited in the ancient lakes before the tilting were initially horizontal and should have also been tilted southward. Of this there is no evidence. What we do see is that the durable Paleozoic and colorful Mesozoic rocks are tilted northeast, as much as four degrees near the south rim of the plateau, decreasing to about one degree toward the interior. But this is part of the saucer shape of the Colorado Plateau, which came about when the saucer shape was formed, not later. The working of erosion upon these tilted beds is what gave rise to the Grand Staircase, so majestically displayed when one drives north from Jacob Lake on the Kaibab Plateau.

The effect of erosion upon beds tilted so long ago may well be part of the reversal story. When erosion attacks a rock couplet that consists of a hard caprock overlying softer beds, the result is a scarp that with time gradually retreats down the dip (maximum slope) of the beds(Figure 2).
Hard-over-soft couplets are common in the Mesozoic section of the Colorado Plateau, and many have produced the remarkable cliffs of the region, including those of the Grand Staircase. We need only mention the Vermilion, Chocolate, and Echo Cliffs, the battlements of Black Mesa, and even the monoliths of Monument Valley to make the point clear (Figure 3). More directly important to our story, the process of cliff retreat has had the overall effect of lowering first the area near the rim of the plateau whereas the saucer’s center once occupied by ancient lakes, remains high the longest (Figure 3). The new effect was a reversal of topographic slope. Thus, we can eliminate the undocumentable idea of back tilting of the strata.
But, is this change in topography enough in itself to cause a reversal of drainage? Not likely, because the formidable mountainous rim of the saucer was still in place and the rivers had to find a way across the barrier to establish a southerly or westerly course. The solution probably lies partly in the ways of rivers, and partly in events west and south of the plateau’s rim.
In a previous Letter, we explored how rivers work, and found that, to erode, they need energy. As we will see in the next Letter, rivers with steep gradients erode vigorously and extend themselves by the process of headward erosion, which enables them to gnaw their way across mountain barriers. Are we starting to hit pay dirt here? Maybe. To find out, we need to expand our views and think about the Land Beyond the Mountains, that shadowy place west and south of the great rim.
The mountains forming the rim did not extend west and southwest forever: somewhere, they came to an end, and what lay beyond them was the sea, not immediately perhaps, but not that far either. Rivers flowing toward the sea worked their headwaters back into the mountains. If the country inland of mountains, namely, our great saucer, was also well above sea level, the rivers would eventually carve their way across not only the mountains, but also into the saucer that lay beyond.
We know that the saucer was near sea level when the mountain building took place, because much of the area was occupied by the Interior Seaway. And the saucer must have been lifted above sea level in order for Hunt’s ghostly river network to form. Consequently, whatever uplift took place at this early stage would have happened between when the sea left, about 65 million years ago, and the time of the river network, about 25 million years ago. We now have a mechanism whereby events taking place far beyond the rim of the Colorado Plateau, combined with erosional processes on the plateau itself, would give us a means for reversing drainage directions on the plateau.
But there is another major disturbance that we must consider. In Miocene time, the area west and south of the plateau was pulled apart by a great episode of rifting, causing the breakup of its ancient mountainous rim. This produced the Basin and Range province so well represented in Nevada, great ranges and intervening basins all aligned roughly north-south, which from space look like an army of giant caterpillars marching determinedly north. When a region is pulled apart, its average elevation is lowered. In the Basin and Range province, the ranges are still high, but many of the basins are not. Streams emptying into these basins would have their mouth at a low elevation and their headwaters at a relatively high one, meaning steep gradients and vigorous headward erosion. This is another mechanism by which streams from the west and southwest could have worked their way into the plateau, reversing its drainage network.
In sum, the proposal is that reversal of drainage on the Colorado Plateau resulted from a combination of events. On the plateau, erosion and down-dip migration of scarps, starting near the upturned rim, lowered the region near the rim. The area near the Colorado Plateau’s center remained high. This caused a reversal of topographic slope. Meanwhile, the plateau and its mountainous rim were probably uplifted, encouraging the encroachment of streams that emptied into distant oceans and which carved their way across the rim. This process was hastened when the mountains forming the western edge of the rim were pulled apart and foundered into what is now the Basin and Range province.
Please note that I refer to all this as a proposal, or in scientific jargon, a hypothesis. Information of various kinds is pulled together to develop an idea of what might have happened, or is likely to have happened. It is a reasonable guess, the best that can be done with the scanty information on hand. Tomorrow, new information may prove this hypothesis right, or wrong, as the case might be. But that is the fun of studying things that are veiled by the passage of untold time.
Dr. Ivo Lucchitta
This is the sixth in a series of “Letters from Grand Canyon” by Ivo Lucchitta that will appear in future issues of the bqr.