Geological Survey Professional Paper 374—I Yampa Canyon in the Uinta Mountains, Colorado

Yampa Canyon, northwest Colorado, was incised in the southern flank of the Uinta Mountain arch by Yampa River.

Modern topographic and geologic maps and aerial photographs of the canyon and vicinity have disclosed unusual features, among which are: natural division of the canyon into three sections; marked change in river pattern from the middle to the lower section; in the middle section, radical differences in topography and geology on its two sides; also in the middle section, but only north of the river, several scallop-shaped erosion surfaces or scars partly rimmed by cliffs and with moderately sloping floors; in the lower section, still different topography and general absence of scars; geographic and geologic relations between Yampa River and the Yampa, Red Rock, and Mitten Park faults.

The observed features give further clues to the origin and development of the canyons and anomalous courses of Green and Yampa Rivers across the Uinta Mountains.

South of and crudely paralleling Yampa Canyon is the Yampa fault and its branch the Red Rock fault, both with downthrow on the north. Apparently the Red Rock fault ends at the Mitten Park fault, which has downthrow on the southeast. Thus the Yampa River joins the Green within a westward-pointing triangular graben between these two faults.

The upper section of Yampa Canyon, nearly 2 miles long, cuts stratigraphically downward through the Weber sandstone and Morgan formation, both of Pennsylvanian age, at right angles to the northeast strike and against the southeast dip of about 12° that mark the end of the Uinta Mountain arch. The river's course is fairly simple, and topography and geology on the two sides are similar.

The middle section of the canyon is nearly 20 miles long; its fall is 333 feet, an average of about 17 feet per mile. Near the point where the middle section begins, the strike of the beds swings sharply to a direction north of west, parallel to the axis of the arch; the dip is predominantly 6° SW. This changed structure extends westward to and beyond the end of the canyon.

The lower section of Yampa Canyon is nearly 24 miles long. Descent of the river surface is 176 feet, an average of less than 7-1/2 feet per mile.

In the middle section the river's course is marked chiefly by open meanders and straight stretches. An exception is a "half-turn" meander, convex southward.

The south wall is prevailingly simple and uniform—a steep slope ending upward in a cliff. Its height and width average a quarter of a mile each, and it follows and fits closely each curve of the river. Its intersection with the adjoining upland is sharp.

Except in the "half-turn district," the north wall is wider, less steep, and of irregular shape. It consists chiefly of adjoining scallop-shaped erosion surfaces or scars. Southward, however, these moderately sloping surfaces end in a steeper slope down to the river, making a break in slope convex upward.

The Untermanns' geologic map of Dinosaur National Monument shows conspicuously that the south wall serves as a formation boundary. The upper part of the south wall and the upland immediately south of it consist of the Weber sandstone, about 900 feet thick, which is loosely cemented and highly jointed. The lower part of the south wall, the north wall, and the adjoining belt of upland expose beds of the next older Morgan formation (except in the "half-turn district" where the Weber sandstone remains). The Morgan is about 1,200 feet thick, of sandstone and limestone; the lower part is more resistant to erosion.

The hypothesis is advanced that the scallops north of the river are "meander-migration scars" formed by the progresive downdip (southward) migration and lowering of early meanders of Yampa River, by an unusual form of homoclinal shifting on more resistant beds in the lower part of the Morgan formation. In the middle section of the canyon five such scars are distinguished; because they differ somewhat from each other they are named and separately described.

The lower section of the canyon differs markedly from the middle section in several ways, the most striking and significant of which are: (a) A notable change in river pattern. Meanders are more numerous and more intricately curved. (b) Topography of canyon walls and of adjoining uplands generally different from the two types in the middle section. Cross profiles are asymmetric but alternating because of interlocking spurs with slipoff slopes. (c) An abrupt change in the relation between river and geology. The canyon is predominantly cut in the Weber sandstone, whose contact with the underlying Morgan is mostly north of the river. (d) An almost complete lack of "meander-migration scars." The exception is the Warm Springs (sixth) scar where the Morgan formation is again exposed to and along the river.

I am convinced, for the following reasons, that at one time the site of the present Yampa Canyon was buried under the Browns Park formation of Miocene(?) age: (a) The thickest deposits of the Browns Park formation, perhaps 1,700 feet of which still remain, were in the southeastern half of Browns Park and its extension to Little Snake River. (b) Continuous outcrops of known Browns Park formation extend to the upstream end of Yampa Canyon on both sides of the Yampa River. (c) The rest of the canyon site is almost surrounded by patches of conglomerate and whitish tuffaceous sandstone of Browns Park lithology. (d) In September 1959 the Untermanns found sands similar to those of the Browns Park at four places between the Yampa River and the Yampa fault.

My hypothesis of canyon origin and development is offered as a chronologic outline involving seven steps.

First step.—After the Uinta Mountain arch was greatly uplifted during Laramide orogeny, it was extensively eroded and the detritus was laid down in the flanking basins and around its southeastern end. Repeated small uplifts accompanied this deposition. In late Eocene or early Oligocene time, arching was renewed and extended southeastward as the Axial Basin anticline. At this time the Yampa, Red Rock, and Mitten Park faults may have begun, but proof of this seems lacking.

Second step.—In middle Tertiary time, uplift virtually ceased, but continuing erosion reduced the mountain mass to mature topography. The resulting surface may have been what A. D. Howard called a pediplane. Along the mountain crest were high residual peaks, between which the erosion surface formed nearly horizontal pediment passes. Northward and southward these passes opened out into a pediment, cut on the upland rocks through retreat of the mountain front and sloping away from the ridge with gradually decreasing gradient. On the south flank this pediment truncated the older, southward-dipping rocks at an angle less than their dip, and also cut across the Yampa and other faults if then existing. The surface was at places rather smooth and at others undulating, with low residual hills. The pediment also wrapped around the southeastern end of the arch.

On this south flank the evidence now remaining near Yampa Canyon may record only a single erosion surface.

Third step.—During the Miocene(?) the widespread and varied Browns Park formation was laid down on the pediplane. At least as far east as Little Snake River the water-borne part of this material is thought to have come from the Uinta Mountains themselves. The basal conglomerate found at many places is prevailingly composed of the reddish quartzitic sandstone of the Uinta Mountain group of Precambrian age; locally, however, fragments of Red Creek quartzite of Precambrian age or of limestone of Mississippian age predominate. The upper, much thicker part of the formation consists partly of white or light-gray sandstone thought to have come from the Uinta Mountain group through leaching of its reddish cement; this was greatly augmented by windborne volcanic tuffs.

Upper beds of the formation presumably overlapped westward up the Browns Park valley and also laterally on its walls; concurrently, sand from residual peaks along the crest washed down into the pediment passes and some of it was carried down the north flank, where it met and mingled with the material rising in the valley. At the same time, some of the sand from the passes, and sand and some cobbles from the crest as the mountain mass retreated, were carried down the south flank, where they filled hollows and blanketed the beveling surface, perhaps to a depth of 200 or 300 feet above the canyon site.

Fourth step.—Chiefly after—but to a small amount during—deposition of the Browns Park formation, the eastern part of the Uinta Mountain arch collapsed, as a graben. Probably this took place in many small movements. Above the site of Yampa Canyon and its environs, a long narrow trough on the surface of the Browns Park was formed thus: (a) In a narrow zone along the Yampa fault (repeated by the branching Red Rock fault), steep north dips, caused by drag, in the Browns Park cover and the truncated older beds that previously dipped southward. (b) North of the drag zone, a zone 4 to 9 miles wide in which the surface of the Browns Park formation was virtually flat in a north-south direction but (through tilt) sloped gently about N. 80° W. (c) Still farther north, to the crest of the ridge, a zone in which the southward, depositional slope of the Browns Park formation remained because the broad central part of the graben had gone down almost vertically.

This trough probably continued, with gradually rising floor, far to the east above and north of the Axial Basin anticline, for the graben movement had extended in that direction, though with diminishing force and effect. I now suspect that, east of Little Snake River, the westward slope of its floor was erosional and depositional, hence original, rather than due to tilting and reversal as we earlier thought.

Fifth step.—Presumably somewhat overlapping the fourth step, during the fifth step the trough began to affect the location and direction of drainage. Streams flowing from the Continental Divide down the depositional slope of the Browns Park formation came together in the graben and, augmented farther on by other streams, were guided westward down the trough as a new Yampa River. At first the river was rather straight, but later it established incipient meanders.

Being wholly in the Browns Park formation, the river at any one time should have had a pattern and gradient essentially uniform throughout. Above the present middle section of the canyon the channel perhaps became located along the outer, northern edges of what are now called the first to fifth scars. Above the present lower section (with a probable meander around the outside of what is now the Warm Springs scar) the river pattern presumably was like that farther upstream rather than one of intricate meanders as today.

At length the river at some point cut through the Browns Park formation to the underlying rocks.

Sixth step.—Superposition commenced when the river's course began to be affected by the differing lithologic composition and structure of the undermass. It is tentatively suggested that the river first cut through the Browns Park formation at the mountainward ends of the meanders curving around the sites of the present first and sixth scars. Initially this caused some decrease in the rate of downward erosion at those points and created temporary baselevels upstream from them. However, only a thin cover of the Browns Park formation then remained elsewhere along the river, hence the undermass was relatively soon reached at all points.

Because of the structure of the undermass, when the river cut through the cover it ran on the Weber sandstone or the Morgan formation. In what is now the middle section it was on the upper beds of the Morgan, except in the "half-turn district" where it was on the Weber. In contrast, in what is now the lower section it ran on the Weber, except for the northward meander around the site of the present Warm Springs scar where it was again on the upper beds of the Morgan. Further development that brought today's conspicuous differences in river pattern must have been influenced chiefly by differences in the way those two formations affected erosion.

The upper part of the Morgan was more resistant than the soft beds of the Browns Park. In that upper part, downward rather than lateral erosion became dominant and cliffs perhaps 200 feet high were cut. Then at the north ends of its meanders the river reached even more resistant limestones in the lower part of the Morgan while elsewhere it was still in upper beds. Direct vertical erosion practically ceased at the points of greatest stratigraphic penetration; but as the tendency to cut the channel down to lower altitudes persisted, least resistance was found in a gradual downdip shifting on top of the older, stronger beds. At first the curving north ends of the meanders were flattened along the strike, and then the meanders themselves slowly migrated, cutting floors that sloped southward and rims whose bases grew lower in that direction. Between the Tepee Hole and Browns Hole scars a conspicuous sharpened spur was developed.

Concurrently, the meanders grew smaller and the river became shorter and of larger gradient. By its constant encroachment against the less resistant upper beds, the south wall was kept narrow, steep, and in conformity with the river's curves.

In contrast, where the river flowed on Weber sandstone it now has a pattern of rounder and more intricate meanders, with asymmetric cross sections and interlocking spurs having slipoffs slopes. I believe that this greatly changed and more complex pattern was developed after superposition began; that it resulted from jointing and erodibility of the Weber; and that the river's length in what is now the lower section became greater and its gradient smaller.

Seventh step.—Late in the canyon cutting some rejuvenation probably took place. Otherwise, it is difficult to explain how and when the more resistant beds in the lower part of the Morgan were breached and a steeper slope was cut near the river, looking like a "valley-in-valley." This suggested process less satisfactorily explains the steeper banks near the river below "treads" in the slipoff slopes on spurs in the lower section; for those "treads" appear to be related to structure and to harder beds in the Weber.

Effect of Mitten Park and Red Rock faults.—An apparent old high-level channel suggests that in its last few miles the Yampa was once farther north than today; that it joined the Green at the east end of the Mitten Park fault; and that the enlarged Green River flowed westward for more than a mile along that fault until it established its course and was able to leave the fault plane and continue farther west on the upthrow side. If this hypothesis is correct, then because of southward dip and of jointing in the Weber, subsequent erosion of new deeper channels may have formed the elongated canyon of the Green around Steamboat Rock and diverted the lower part of the Yampa to its present junction with the Green.