USGS Logo Geological Survey Bulletin 1191
Black Canyon of the Gunnison: Today and Yesterday


More specifically, this heading should read "How the canyon is being carved," for the process of cutting is still going on at the present time. And to say blandly that the river is doing the job will satisfy few inquirers, although the river is certainly the driving force. Anyone who has seen the canyon during the high-water stage of early summer can attest to the unleashed fury of the river at that time. Impressive from the rim, the river is almost frightening close at hand. The river is the prime mover of material and the chief agent of erosion, but it is supported in its work by several subsidiary agents. Gullying, rill wash, frost action, atmospheric weathering, and mass wastage of debris under the influence of gravity all tend to widen the canyon walls and transport rock material to the canyon bottom. Material delivered to the canyon bottom is then worn away and carried off by the river, which at the same time erodes its bed with the material in transit.

The work of the river is mainly seasonal—during high water, when it is swollen by the snowmelt of its 4,000-square-mile drainage basin and when its ability to move a suspended load is high—whereas most of the other processes are nonseasonal, frost action excepted. Frost action of course is primarily a wintertime process. It takes advantage of such natural weaknesses in the rocks as fractures and foliation planes. Water seeps into these places, freezes, and slowly forces an opening. Then in a ratchet-like action, repeated freezes and thaws force it wider.

Water is essential also in the slow chemical breakdown of rocks by atmospheric weathering. This process, too, is most effective along lines of natural weakness where moisture can gain access.

Gullying and rill wash take advantage of natural zones of weakness in the rocks, and most of the tributary ravines in the national monument section of the canyon utilize such zones. Joints, faults, foliation planes, and even diabase dikes become loci for natural drainage—diabase dikes, not because of structural weakness but because of rapid chemical breakdown under the influence of weathering.

Through the widening of joints and fractures, the physical stability of large rock masses on the steep canyon walls is reduced ultimately to the point of failure. Blocks then plummet, bounce, or creep to the canyon floor where they are attacked and worn away by the river. The great size of many of these blocks bespeaks the effectiveness of the process. Helter-skelter blocks the size of houses, looking like stepping stones from the canyon rims, pose a challenge to venturesome hikers along the canyon floor.

Blocks in excess of what the river can handle in a given time form aprons of talus against the canyon walls (fig. 4).

FIGURE 4.—Talus blocks in canyon bottom at foot of Painted Wall, August 29, 1962. Foot passage through such areas is precarious. River is flowing at a rate of about 250 cubic feet per second. Falls at left are about 10 feet high. During high water, all the large blocks in the foreground are engulfed in a white-water rapid.

The abundance of talus on the lower slopes, as shown by figure 4, suggests a short-term imbalance between the effectiveness of subsidiary agents of erosion and the capacity of the river, an imbalance destined to increase as water is impounded behind dams upstream. But the great depth of the canyon relative to the width leaves little doubt of the long-term competence of the river—in the long run the river is lowering its bed faster than the rims are receding.


Widespread evidence in the canyon country of the Western United States indicates that rivers once firmly entrenched are little able to alter their courses, despite seemingly anomalous relations to rock hardness and structure. The Black Canyon is an outstanding example of this very principle, as the early geologists who visited the area quickly recognized. Flowing quite normally between highlands north and south in its upper reaches, the Gunnison River strikes through the very heart and heights of the Gunnison uplift in the depths of the Black Canyon. Here it cuts deeply into the hardest of rocks, seemingly aloof to easier possible courses through softer rocks north and south.

But the present course of the river is controlled by its initial course at the time downcutting began—not by the rocks through which it now flows. (See fig. 7.) Down-cutting began when the river flowed across a terrane of newly erupted volcanic rocks in late Tertiary time. These rocks were deposited sheetlike without regard to underlying structure. Partly eroded remnants still cap the mesas near the head of the canyon. Once entrenched into these rocks, the river had no alternative but to pursue its established course. Its chance superposition across the hard Precambrian rocks of the then-buried Gunnison uplift was a lucky event that made the Black Canyon possible.


Another critical factor joined the interplay of coincidences that made the Black Canyon possible. In its passage through the canyon, the Gunnison River drops more than 2,100 feet. This exceptionally steep gradient affords the river the energy needed to incise its bed downward at a rate higher than subsidiary agents of erosion can flare out the rims. The result is a very narrow canyon, deeper in many places than wide.

During high water, the Gunnison River usually has a peak discharge of about 12,000 cubic feet per second through the Black Canyon. At this discharge the river expends energy at a rate of more than 1-1/2 billion foot pounds per second, or about 2-3/4 million horsepower, through the 50-mile-long stretch of the canyon. Converted to electricity this energy would exceed 2 million kilowatts.

Most of this energy is dissipated by turbulent flow—the frictional churning of the water dashing over rapids and against rocks in its wild descent through the canyon. Part of the energy is expended in scouring the bed of the river, and the remainder is consumed in transporting debris. After construction of the new hydroelectric facilities at Blue Mesa and Morrow Point, a fraction of this energy will be harnessed and converted to electricity.

The steep gradient of the river is dependent on physiographic controls downstream from the canyon in the lower Gunnison valley and even along the main stem of the Colorado River. These controls are partly structural, partly lithologic, and partly purely hydrologic. Entrenchment began as soon as they became effective.

Much impetus was added to the cutting of the Black Canyon by a spectacular physiographic event far downstream in the Grand Junction area, recently described in a paper by Lohman (1961, p. B144). In brief, the combined flow of the Gunnison and Colorado Rivers once crossed the Uncompahgre Plateau in what is now Unaweep Canyon. Unaweep Canyon was abandoned, and flow was diverted around the north end of the plateau in a remarkable sequence of drainage adjustments. As a direct consequence, the energy and competence of the entire upper Colorado River system were vastly increased.

Just below the mouth of the Black Canyon, the upper Gunnison (main stem), the North Fork, and the Uncompahgre merge into a single large river. While the upper Gunnison was cutting the Black Canyon in hard Precambrian rock, the lower Gunnison—rejuvenated by the abandonment of Unaweep Canyon—was excavating a capacious valley in soft Mancos Shale. The lower Gunnison, the North Fork, and the Uncompahgre, moreover, were all lowering their beds more rapidly through soft Mancos Shale than the upper Gunnison was lowering its bed through hard crystalline rock; the net result was steepening of the gradient of the upper Gunnison. If rock conditions were alike both upstream and downstream, therefore, there would be no Black Canyon.


Basically, two interacting factors are responsible for the steepness of the canyon walls—first, the extreme resistance of the rocks to erosion, especially the granitic rocks (fig. 5), and second, the high efficiency of the river in eroding its bed as opposed to the low efficiency of subsidiary agents of erosion in widening the walls. But why haven't comparable rivers cut comparable canyons through similar rocks in other parts of Colorado? Nearly identical rocks are widespread in Colorado, and many rivers have cut deep canyons through them. Just 50 miles to the east, for example, the Taylor River has cut a great slash through similar if not identical rocks in its descent from the Sawatch Range. The Colorado cuts similar rocks in Glenwood Canyon, and so does the South Platte in the Front Range. Many other examples could be cited. But none of these valleys has both the depth and verticality of the Black Canyon. The Royal Gorge of the Arkansas is perhaps most comparable, but it is far overshadowed by the Black Canyon.

Why then is the Black Canyon unique? Rapid down-cutting alone is not the answer. Downcutting must be combined with resistance on the part of the valley walls to forces tending to flare them out. While the Gunnison has cut its Black Canyon the adjacent Uncompahgre has eroded a valley equally deep but 20 to 40 times as wide. The nonresistant sides of the Uncompahgre Valley have enlarged 20 times as fast laterally as vertically.

In most valleys, gullying and rill erosion by tributary streams keep pace with downcutting by the master stream. Such tributaries respond effectively to changes in regimen of the master stream. If the master stream lowers its bed, so do the tributaries because their gradients are steepened and their energy is increased. Conversely, if the master stream aggrades, the tributaries do the same as their gradients are flattened and their energy is reduced. Major tributaries of the Gunnison such as Lake Fork, Curecanti Creek, Blue Creek, Smith Fork, and Cimarron Creek are fairly well in accord with this rule. All of them are sizable perennial streams having large catchment basins and reliable discharges.

FIGURE 5.—Chasm Wall, at north end of Vernal Mesa, has a nearly vertical drop of 1,800 feet. Only the upper half of the cliff is shown in this view. Rock is Vernal Mesa Quartz Monzonite. Exfoliation along vertical joints contributes to sheerness.

But minor tributaries such as Grizzly Gulch flow intermittently from woefully small catchment basins. They terminate precariously, therefore, at the canyon rims or they find their way to the river down precipitous ravines along lines of weakness in the rocks. In either event, their erosive power is negligible compared to that of the Gunnison.

The small size of their catchment basins is due chiefly to the form of the Gunnison uplift. Standing as it does above the surrounding country, the uplift sheds most of its precipitation away from the canyon to such streams as Smith Fork, Crystal Creek, and the Uncompahgre River. Many small streams that formerly flowed to the canyon, more over, have been diverted by natural drainage adjustments into streams that flow away from the canyon. As a result, only the shortest, least competent tributaries flow directly to the river in the depths of the canyon. These tributaries simply cannot keep pace with the Gunnison.


In attempting to evaluate the age of the Black Canyon, the geologist must be mindful of many variables and many uncertainties. First of all, he is not dealing with precise figures but with relative ages covering an incomprehensible immensity of time (fig. 6). In terms of relative geologic age, viewed in the vastness of geologic time, the canyon is very youthful indeed—it formed only yesterday. But in terms of man's experience, his short recorded history, and his even shorter life span, the antiquity of the canyon is staggering.

FIGURE 6.—Major divisions of geologic time.

The long chain of events that led up to the actual cutting of the Black Canyon began to take form in the dimmest recesses of the past, well over a billion years ago during the Precambrian Era. Much of geologic time elapsed while the stage was being set. The really critical interplay of circumstances began about 60 million years ago, at the time of the Laramide orogeny, when the Gunnison uplift first took form. These events are further described on the pages that follow. (See also fig. 7.) But the actual cutting of the canyon started perhaps a scant 2 million years ago.

Acknowledging the many gaps in the local geologic record, we can, nevertheless, place fairly close limits on the time when cutting could have started. First of all, the Gunnison River did not incise its canyon until volcanism had died out in the nearby West Elk and San Juan Mountains (fig. 7). Therein is one of the real clues to the mystery. The youngest volcanic rocks that clearly antedate the canyon belong to a sequence of dark basaltic lava flows called the Hinsdale Formation. Scattered remnants are all that remain of this formation near the Black Canyon, although large masses are preserved in the San Juan Mountains. The age of the Hinsdale is generally, but doubtfully, regarded as Pliocene, an epoch that began about 11 million years ago and ended about 1 million years ago.

FIGURE 7.—Critical stages in the formation of the Black Canyon landscape. (click on image for an enlargement in a new window)

A. Late Laramide stage (early Tertiary).—The newly formed Gunnison uplift (1) and the Sawatch Range (2) are being attacked by erosion. Initial drainage is north to a large lake (3) centered in central western Colorado and eastern Utah.

B. Planation stage (middle Tertiary).—Erosion has planed off the Gunnison uplift and has greatly reduced the Sawatch Range. Streams course freely across a broad flat plain, independent of underlying rock structure.

C. Volcanic stage (middle to late Tertiary).—Repeated volcanic eruptions have built up mountainous piles of debris in the newly farmed West Elk Mountains (1) and in the San Juan Mountains to the south (2). Drainage has been diverted south around the periphery of the West Elk Mountains across the buried Gunnison uplift.

D. Present stage.—Gunnison River has entrenched itself into Precambrian core of exhumed Gunnison uplift, and in so doing has carved Black Canyon. (1) Shoulder of Grand Mesa, (2) North Fork, (3) Uncompahgre River at Montrose, (4) Black Canyon, (5) West Elk Mountains, (6) Lake Fork, (7) Gunnison River at Gunnison (8) Monarch Pass.

After the Hinsdale was erupted, it was attacked by erosion, and its debris was redeposited as bouldery gravel. This also was a Pliocene event that preceded canyon cutting. Remnants of the gravel still cap some of the mesas near the head of the canyon. After Pliocene time, during the Ice Age or Pleistocene Epoch, glaciers repeatedly occupied the high valleys of the nearby mountains. Melt waters from the glaciers deposited terrace gravels along the Gunnison and its tributaries at altitudes well below the rims of the Black Canyon. Evidently then, canyon cutting began well before the start of glaciation in Pleistocene time; it must have begun late in the Pliocene. The best guess based on present knowledge places the event about 2 million years ago. Therefore, in a canyon that averages about 2,000 feet in depth, the rate of downcutting has been about 1 foot per 1,000 years.

mountain peak

<<< Previous <<< Contents >>> Next >>>

Last Updated: 28-Mar-2006