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


The Yampa River, which rises in the Park Range, was described by Hancock (1915, p. 184), who then fully analyzed the development of its middle course across the Axial Basin anticline and Juniper and Cross Mountains by superposition from the Browns Park formation.

The present report deals primarily with that part of the river downstream from Hancock's area.



Yampa Canyon begins at point A (see pl. 1) where the river, after crossing the low ground at the mouths of the Vale of Tears and of Disappointment Creek, cuts into southeastward-dipping Weber sandstone. This is about 0.7 mile upstream from (southeast of) the point where the river crosses the eastern boundary of Dinosaur National Monument.

The canyon ends at the mouth of Yampa River (point D), where it joins Green River just east of Steamboat Rock.

From the upper end of the canyon to its mouth the airline distance, in a direction N. 78° W., is about 24-1/3 miles; but because of the meandering course the distance by river is nearly twice as great, or about 45-1/6 miles.1

1River distances in miles, beginning at the mouth, and altitudes of the river surface are shown on the "Plan and Profile of Yampa River, Colorado," from Green River to Morgan Gulch, 1924, U.S. Geol. Survey.

At the upper and lower ends of the canyon (points A and D) the altitudes of the river surface are respectively about 5,589 and 5,064 feet; the river thus falls 525 feet within the canyon, an overall average of more than 11.6 feet per mile.

The maximum depth of canyon noted is about 1,715 feet; this is at a point opposite Warm Springs Draw, 4 miles upstream from Green River, where the Yampa surface is at 5,085 feet and the top of Warm Springs Cliff on the south side (less than 200 yards horizontally away from the river) is at about 6,800 feet.


Crudely paralleling Yampa Canyon, but south of Yampa River at all points, is the Yampa fault. The Untermanns (1954, p. 151-152) describe it as the largest fault in Dinosaur National Monument, of the normal type with its fault plane dipping to the north at angles of 50° to 75° They add, "East of Johnson Draw, at the foot of Tanks Peak, Precambrian (Uinta Mountain group) occurs against lower Triassic (Moenkopi), producing a vertical displacement of between 3,600 and 4,000 feet, maximum for faults of the Monument area."

The geographic and geologic relations of the Yampa fault to the ancient and present courses of Yampa River had a significant bearing on the views developed by Bradley, Gilluly, and me, and presented (Sears, 1924a) as a part of our general concept. Those relations, now known with more details and more certainty than in 1922, form an essential base for the hypothesis herein presented.


As briefly described by the Untermanns (1954, p. 152), "The Yampa fault has several branches; the largest, which the writers have called the Red Rock fault, begins at Red Rock Draw and runs in a northwesterly direction beyond Pool Creek where it possibly intersects the Mitten Park fault." The downthrow side is to the northeast.

From their map the Red Rock fault has been drawn on plate 1, herewith. Its junction with the main Yampa fault, the repetition of beds that it caused, and the steep northeastward dips on both faults, due to drag, are all conspicuous on aerial photographs.


As mapped by the Untermanns, the Mitten Park fault follows a generally northeastward but curving course. Downthrow is on the southeast side; where the fault crosses Green River downstream from Steamboat Rock, the displacement as estimated by the Untermanns (idem, p. 154) is between 1,500 and 2,000 feet. At the north end of Steamboat Rock the beds on both sides of the fault are greatly steepened by drag. To the east, in the canyon of Green River up stream from its junction with the Yampa, the fault appears to die out sharply and turn into a flexure whose magnitude diminishes eastward.


A glance at plate 1 shows that between the Red Rock and Mitten Park faults is a graben or structurally depressed area with the shape of a westward pointing triangle. This is perhaps the feature to which Powell (1876, p. 202 and pl. 5) referred as the "Echo Park sag."

This triangular graben is added upon and accentuates the depression effect of the major graben to which Powell (1876, p. 209) called attention with the theory that after Browns Park deposition the eastern end of the Uinta Mountain arch collapsed. Because of the concept that Bradley, Gilluly, and I formed in 1922, the collapse and the major graben were discussed at length and then summarized in the following words (Sears, 1924a, p. 291-303):

"The collapse was caused by a single large fault [the Yampa fault] on the south, by flexures and distributive faulting on the north, by tilting and some faulting on the east, and by tilting on the west."

The Yampa River flows in this major graben, near and roughly parallel to its southern margin, is therefore not a new idea. Because of the later and more detailed mapping by the Untermanns, however, I wish to emphasize that in its lower course Yampa River runs into, and joins Green River within, the added depression or triangular graben between the Red Rock and Mitten Park faults—a complicating problem to be discussed under the last heading of this report.


As mentioned in item 1 of the Introduction, the 45-mile Yampa Canyon is naturally divisible into three parts—a short upper section, and a middle and a lower section of roughly equal length.

It is therefore both logical and convenient to divide the detailed description of the canyon under separate headings for those three sections.


The upper section of the canyon, as herein designated, extends from point A (pl. 1), the beginning of Yampa Canyon in sec. 20, T. 6 N., R. 99 W., downstream about 1-2/3 miles to point B, just southwest of the high, sharp westward-jutting spur north of the river in the SE. cor. sec. 18, T. 6 N., R. 99 W. Within this section the river's course is rather simple, forming an almost straight line in the upstream half and three small open meanders in the downstream half. In contrast to the longer middle and lower sections of the canyon, this short upper section is noteworthy in that the topography and geology on the two sides of the river are so nearly identical.

Topographically, cross sections of the canyon are almost symmetric; small differences in the angle of slope of the two walls suggest, however, that the meanders may have been slightly enlarged during incision.

Geologically, the upper parts of both walls and the upland behind them expose Weber sandstone, and the lower parts of the walls expose beds of the Morgan formation. (See pl. 1; figs. 1, 2.) As the beds here strike northeastward and dip about 12° SE., the river in its general northwesterly course flows at right angles to the strike and against the dip, cutting stratigraphically down from the top of the Weber into the lower part of the underlying Morgan.

FIGURE 1.—Entrance to Yampa Canyon; view downstream. Dip slope is on Weber sandstone. (National Park Service photograph.)

It should be emphasized, however, that structurally the beds exposed in this upper section of the canyon represent only the lower part of a wider zone of southeastward-dipping rocks whose stratigraphic sequence along the river is shown by the Untermanns (1954, pl. 2) to include a dozen formations. At some point in Lily Park (perhaps near the junction of Little Snake River with the Yampa, several miles upstream from and east of Dinosaur National Monument and the east edge of the Untermanns' map) the soft lower beds of the mile-thick Mancos shale of Cretaceous age begin the southeastward dip and northwestward rise that here mark the southeast end or nose of the Uinta Mountain arch (Sears, 1924b, pl. 35). From that point the Yampa cuts downward into succesively older beds, at last reaching the Weber sandstone and the Morgan formation, whose relative hardness, thickness, and attitude have permitted the erosion of a deep continuous canyon.

FIGURE 2.—Upper section of Yampa Canyon; view upstream. Upper part of Morgan formation; probably some Weber sandstone at top in distance. (National Park Service photograph.)


The middle section of the canyon, as herein designated, extends from point B downstream for about 19-2/3 miles to point C at the mouth of Big Joe Draw in Starvation Valley. (See pl. 1.) The altitudes of the river surface at points B and C are respectively 5,573 and 5,242 feet; thus in this section the Yampa falls 333 feet, an average of more than 16.9 feet per mile.

In the vicinity of point B, the structure of the rocks in and on both sides of the canyon begins to change in a pronounced manner. From the northeastward strike and southeastward dip that characterize the upper section, the strike swings rather sharply to a direction somewhat north of west (ranging approximately from N. 65° W. to N. 75° W.) and the dip is prevailingly 6° SW. (with an observed range of 3° to 10°). This changed structure, with these strikes and dips, extends westward to and beyond the junction of the Yampa with the Green; southward from Yampa River, generally for 1 to 3 miles, until the Yampa fault is approached; and northward for a number of miles as a part of the south flank of the Uinta Mountain arch.

The north-of-west strike just described is fairly close to the overall direction of flow of Yampa River which, as previously stated, is N. 78° W. for the airline from point A to point D.


Within the middle section the course of the river is marked by large- and medium-sized meanders interspersed with a few almost straight stretches a mile or more in length. With a single exception, all the meanders are of the open type. The exception is the meander in secs. 22 and 27, T. 6 N., R. 100 W., which is convex southward and is about half a mile long and three-tenths of a mile wide; it is of the type called by Davis (1914, p. 23-24) "half-turn." Because this meander and its environs north of the river are exceptional in several other ways as well, they will be mentioned and discussed repeatedly; for brevity, these environs will be referred to in this report as the "half-turn district."



The south wall of Yampa Canyon is very simple and uniform. Except for a few short reentrants where interrupted by side streams, the wall is virtually continuous as a steep slope ending upward in a sheer cliff. The height and width of the south wall in this section average roughly a quarter of a mile each. At places (particularly in sec. 20, T. 6 N., R. 100 W., and in secs. 13, 14, and 15, T. 6 N., R. 101 W.) the cliff at the top is complicated by very small crenulations, but as a whole it is simple. Thus the cliff and slope follow and fit into each curve of the river with noteworthy preciseness.

Almost without a break, the 6,000-foot contour extends along the upper part of the south wall, at varying distances below its top (which ranges in altitude from about 6,300 to about 6,875 feet).

The intersection between the canyon wall and the upland adjoining it is sharp and nearly at right angles, and shows little if any trace of rounding by erosion. Indeed, many knolls and larger hills on the edge of the present upland are partly sheared by the cliff.


Except for about 3-1/2 miles along the river in the "half-turn district," the north wall differs radically from the south wall, particularly in being much wider, much less steep, and of very irregular shape. Its width, in contrast to the nearly uniform quarter of a mile for the south wall, averages approximately a mile and ranges from about 3/4 to 1-1/2 miles. Only in a general way do the bends of its upper rim correspond to the present curves of the river.

As the greater width, more moderate slope, and irregular shape of the north wall are related to what are herein designated as meander-migration scars, they will be more fully discussed under a heading dealing with those scars (p. I-9).

Another feature of the north wall is of geomorphologic significance. Southward the moderately sloping floors of the scars are terminated by a much steeper slope down to the river, making a convexity upward. This break in slope is conspicuous on the aerial photographs. However, conditions here are somewhat anomalous and puzzling. The photographs show tonal and other differences suggesting that the break in slope is related to some variation in the resistance of rock layers. (The horizon of tonal and presumably lithologic change is discernible also across the river; but the south wall is in general so narrow and steep that at only a few spots is there even a faint trace of any break in slope.) This apparent relation between break of slope and stratigraphic horizon seems to be borne out by two other observations: (a) around each meander the steeper slope looks to be a little wider horizontally and a little higher vertically northward updip; and (b) as a whole the steeper slope is somewhat wider horizontally and higher vertically downstream as the river falls.

The rim or sheer cliff that caps the north wall has some resemblances to and some differences from the one that caps the south wall. Along its top the lowest points, like those on the south wall, are at an altitude of about 6,300 feet. The highest points, however, reach an altitude of about 7,250 feet, as contrasted with a maximum of 6,875 feet on the south. The north rim is less continuous, being interrupted at the north ends of the meander-migration scars (as discussed later). However, where it exists, the north rim resembles the south rim in its sharp angle of intersection with the upland surface behind it, and in its abrupt shearing through knolls and larger hills on that surface.



The upland south of the middle section of the canyon is rather level and smooth on its eastern half, forming areas called East Cactus Flat and West Cactus Flat.

Farther west, the surface of this upland is much more irregular. Here and there, altitudes are a little higher near the canyon and tend to be somewhat lower within 1 or 2 miles to the south, beyond which they rise fairly steadily to the Yampa fault and Blue Mountain behind it. With this southward rise of the surface, the 7,000-foot contour lies at or south of the Yampa fault as mapped by the Untermanns, except for a stretch of about 1-1/2 miles in secs. 28 and 29, T. 6 N., R. 101 W., north of Tanks Peak. (See pl. 1.) Because this exception, if valid, seems here to present an anomalous relation between structure and topography, the aerial photographs of this vicinity were scrutinized with extra care. These photographs give some indications that a second fault exists here parallel to and about half a mile north of the Yampa fault as mapped; and that it compares and perhaps even connects with the northern of the 2 faults 6 miles farther east.

The irregular topography of the upland between Yampa Canyon and the Yampa fault is accompanied by a no less irregular drainage pattern. Most of the numerous streams (all shown as intermittent) begin on Blue Mountain and extend northward across the Yampa fault for a couple of miles. Thereafter they assemble in a few principal channels which (for no reason obvious on the topographic map, but perhaps related to the jointing in the Weber sandstone) follow pronounced eastward or north-of-westward courses for 2 to 5 miles until a further swing allows them to find their way to the river. To these middle courses are added a few short southward-flowing streams. Especially noteworthy are two that begin at the very edge of the upland above the canyon wall, flow southward and then together eastward until joining the stream in Dry Woman Canyon just before it slips down the steep south wall and into Yampa River.


The upland north of the canyon differs topographically from that on the south in several ways. For one thing, it is substantially higher. As already described, the south upland between the canyon and the Yampa fault (with the questionable exception of a small stretch north of Tanks Peak) is lower than 7,000 feet in altitude. In contrast, the 7,000-foot contour line north of the Yampa crudely parallels the river a couple of miles away and also surrounds many headlands and hills in the zone southward to the canyon wall. Still higher land lies to the north, on the flank of Douglas Mountain; a short fragment of the 8,000-foot contour line is seen on the map at the head of Buck Draw, in sec. 16, T. 7 N., R. 101 W.

Between the main 7,000-foot contour and the outliers of that contour around hills and headlands to the south is a zone of somewhat lower altitudes. A number of the longer streams, which prevailingly rise on the high flank of Douglas Mountain and flow southward to the Yampa River, have distinct, short or long, right or left bends in their middle courses where crossing this lower zone, in a manner suggesting stream piracy. The outliers have a general cuesta form; their upper surfaces, eroded into somewhat steplike topography, show southward gentle dip slopes on several beds; whereas their northern, northwestern, and northeastern edges are steep slopes or escarpments cut downward across the dip to the vale (the zone of lower altitudes described above) and face the main 7,000-foot contour to the north, which lies on the southern dip slopes of still older beds. This feature is particularly well shown in the east-west ridge in sec. 11, T. 6 N., R. 100 W.

Still farther north, the older formations below the Morgan are cut by the streams in such a way that they tend to form flatirons dipping gently southward and pointing northward.


Perhaps the most conspicuous and hence first-noticed feature shown on the Untermanns' map (1954, pl. 2) is the way in which (within what is here termed "the middle section") the south wall of Yampa Canyon serves as a formation boundary. This feature cannot be fortuitous. Not only is it one of the criteria by which the middle and lower sections of the canyon have been differentiated, but also it is intimately related to the geomorphology of the river.


Except for a very short distance in sec. 27, T. 6 N., R. 100 W., where the top of the south wall of the canyon near the "half-turn" meander is now cut back to the Yampa fault, the upland adjoining the canyon in a belt of varying width is mapped as wholly developed in the Weber sandstone, dipping about 6° a little west of south.

The Untermanns (1954, p. 36) describe the Weber as

a uniform, well-sorted, buff to white or gray, medium- to fine-grained quartz sandstone. * * * Most of the cementing material is calcareous although it becomes quartzitic locally. * * * The poorly cemented and highly jointed nature of the Weber accelerates its erosion, producing characteristic deep steep-walled gorges and resulting in extremely rough topography.

They add that the thickness of the Weber sandstone in the eastern portion of Dinosaur National Monument is 850 to 900 feet.

The boundary of the Weber with the underlying Morgan formation lies almost continuously high up along the south wall of Yampa Canyon. Because of the steepness or verticality of the upper part of that wall, and because of blurring in black-and-white reproduction of the Untermanns' topographic-geologic map, it was impossible to determine at each point precisely the altitude of the contact or the thickness of Weber sandstone that now remains above that contact at the edge of the upland; fortunately, however, these details are of little if any significance for the problems herein discussed.

From the foregoing description it follows that the lower, major part of the south wall of Yampa Canyon throughout the middle section exposes beds of the next older Morgan formation.


The north wall of Yampa Canyon and (again with the exception of the "half-turn district") the belt of upland adjoining it are mapped as wholly developed in the Morgan formation, next older than and dipping under the Weber sandstone south of the river.

According to the Untermanns (1954, p. 33-34):

The contact between the Weber and Morgan formations was placed at the base of the massive Weber sandstone and at the top of the first limestone bed below it. * * * The sandstone beds in both formations are very similar, consisting of uniform fine-grained quartzitic to calcareous quartz-sandstones. The upper part of the Morgan appears to be transitional into the Weber. The light buff to gray color of the Weber is characteristic of the upper sandstone beds of the Morgan, although both formations contain some red sandstones. * * *

The upper third of the Morgan consists of thin layers or compact, often very cherty, gray limestones which weather red. They alternate with thick fine buff to terra cotta-colored sandstone beds, occasionally somewhat cross-bedded * * * which may exceed 100 feet in thickness.

In their measurements for the Hells Canyon area (a few miles farther west, near the middle of what is herein termed "the lower section of Yampa Canyon") the Untermanns (idem, p. 160-161) give a thickness of approximately 1,200 feet for the Morgan formation.


Again with the exception of the "half-turn district," the entire north side of the middle section of Yampa Canyon from point B to point C is made up of adjoining large scallops, each of which is partly rimmed by cliffs or very steep slopes and has a floor that descends with moderate slope nearly to the river.

These scallops are herein referred to as meander migration scars because they are believed to result from and record the progressive downdip (southward) migration and lowering of early meanders of Yampa River.

The unqualified term "meander scar" has apparently been used but rarely in the literature, and then (whether the meanders are on flood plains or are incised) only with expressed or implied reference to the trace left as an oxbow after a cutoff. (For example, see Thornbury, 1954, p. 130-131.)

Cotton (1949, p. 250) uses the term "meander-scar" as an adjective qualifying alternate terraces developed during side-to-side swinging of a meander belt.

These meanings and applications are mentioned here to emphasize that the term "meander-migration scar" is intended to have a quite different meaning (which, though partly anticipating suggested explanations offered later in this paper, is indicated at this point for convenience).

The cliffs that are so conspicuous on the sides of these scars are now interrupted and absent at their upper or inner (north) ends. I find on aerial photographs and on the topographic map no conclusive evidence as to whether or not the cliffs once were almost continuous. However, I am disposed to think that they were (though probably low at their upper ends); and that later their northern parts were dissected and obliterated by the streams which, rising on the flanks of Douglas Mountain, flowed southward farther and farther to join the migrating river.

As far as I know, these meander-migration scars, all on the north side of Yampa River, have not hitherto been observed or at least have not been mentioned in a published statement. Indeed, rather wide reading, search of maps, and conversations have not brought to my attention any good example of such an extensive feature elsewhere or any clear and specific description of the feature or of the process by which it evolved.

As the meander-migration scars in the middle section of Yampa Canyon differ from each other somewhat in size, shape, and other ways, they are separately but briefly described below.


The Anderson Hole (first) scar begins at point B, which has been selected as marking the boundary between the upper and middle sections of the canyon. On the east the sear adjoins the upland lying north of the upper section, where the beds dip toward the southeast. On the west it adjoins the "half-turn district," where the upland includes a substantial outcrop of Weber sandstone—the only remnant of that formation north of the middle section. This scar, with its rimming cliffs and low inner floor, forms a protected hollow that is known as Anderson Hole.

The conspicuous cliffs that face each other on the east and west sides of the scar are almost 2 miles apart near the river and almost 1-1/2 miles apart at their present north ends. In this instance the south ends of both cliffs are very close to the river. Northward the tops of the cliffs rise in altitude, whereas the cliffs themselves become lower and gradually turn into steep slopes. (The northward-facing steep slopes east of point M and west of point N on plate 1 are believed not to be part of the scar rim described above, but to be a product of the later cuesta development discussed on page I-9.) I picture these rimming cliffs or steep slopes as once extending farther and connecting in a gentle curve that formed the low north rim of the scar, later worn down and obliterated by the streams that now cross its site. (Admittedly, however, the former existence of such north rim, as well as its location, height, and degree of continuity, seem to be questions for deduction and not susceptible of proof. As a collateral question: if such a north rim existed when a meander of the river was at that altitude and position, were there then three tributaries flowing into that meander, and were those tributaries extended as the meander migrated southward down the dip? or did the tributaries begin at some later stage? I would lean to the first alternative.)

The floor of the scar, between the east and west cliffs and the hypothetical north rim, now has considerable relief through dissection by present streams, but overall it has a moderate slope southward nearly to the river, dropping some 1,100 feet in about 1-1/2 miles. An interesting feature, discernible on the topographic map but more conspicuous on aerial photographs, is a low concentric supplementary rim and part of a second, in the middle of the scar. Only partial traces of similar supplementary rims are detectable in any of the other scars.


The Tepee Hole (second) scar, whose floor forms what is called Tepee Hole, is sufficiently like the Anderson Hole scar that most of the same description would apply. The chief differences are as follows: (a) The east cliff (rather than the west) adjoins the "half-turn district"; and the south end of this east cliff is close to the river. (b) The south end of the west cliff is nearly half a mile from the river. (c) Between the west cliff of the Tepee Hole scar and the east cliff of the Browns Hole scar is an upland that terminates in a spectacular southward-pointing sharpened spur. (See fig. 3.) (d) The position of the hypothetical north rim is less clear than that in the Anderson Hole scar for it depends on whether the east cliff should be considered to reach point Q (pl. 1) or whether the part northeast of point P is instead the product of, or has been modified by, the later cuesta development discussed on page I-9.

FIGURE 3.—Sharpened spur between Tepee Hole (second) and Browns Hole (third) scars. View northward across canyon. West Cactus Flat and Haystack Rock in foreground.


Browns Hole is the name given to the sloping floor of the third scar. In many ways this scar resembles the two farther up the river, but it differs from them in other ways that are of interest and significance. For one thing, in horizontal plan the scallop is much shallower—that is, its north-south distance is much shorter than that from east to west. Even more significant, its rimming cliff is much more continuous and complete, being broken only at its northern curve by the gorge (about 1,500 feet wide) through which flows this stream that drains Browns Draw and on its northwest side by an even narrower gorge through which flows the unnamed stream, one of whose upper branches drains Iron Mine Basin. The greater continuity and the curve of this cliff seem to me to illustrate and support the concept, discussed above, that the Anderson Hole and Tepee Hole scars also once had north rims.

The stream that cuts through the cliff on the northwest side is itself unusual; for of all the streams that interrupt the side cliffs in all five scars, it is the only one that flows for any considerable distance on the upland before reaching the cliff. However, its unusual length was perhaps not original but caused by piracy; this is suggested by the sharp bend of the stream (elbow of capture?) about 1 mile northwest of the gorge and by the lowness of the divide between that bend and the west fork of Browns Draw.


The Bower Draw (fourth) scar (which may be identified by the name of the principal channel, Bower Draw, that crosses it) is much less distinct. Indeed, its size and shape are such that its nature might have gone unsuspected had not the other scars been noticed and analyzed. Perhaps it might more logically be divided into 3 merging scars—2 short and very shallow ones at the east and a larger one up Bower Draw at the west; but this would seem to be an undesirable complication. The general continuity of its cliff, the moderate slope of its floor toward the river, and the approximate accordance of its cliff pattern with the present curves of the river—all together appear to me to be ample evidence that it too was formed by the lateral downdip migration of meanders. I think that the difference was caused by the greater straightness and lack of large meanders in the early as well as the present course of the river through most of this stretch.


The Five Springs Draw (fifth) scar (identifiable by Five Springs Draw, which crosses it) is the last scar in this section of the canyon. The southwest end of its northwest cliff is close to the river near point C, which has been selected as marking the boundary between the middle and lower sections of the canyon.

The Five Springs Draw scar is much like the Anderson Hole and the Tepee Hole scars, and therefore will not be described in detail. However, it may be well to emphasize that the aerial photographs of the Five Springs Draw scar as well as those of the Bower Draw scar show clearly the break in slope and upward convexity near the river, as discussed on page I-8 under "North wall."


At several places, mention has been made of ways in which the "half-turn district" differs sharply from the rest of the north side of the middle section. As these differences are thought to be very significant, for convenience they are assembled and repeated here in a single place, as follows: (a) The "half-turn" meander contrasts with the "open" type of meander and the almost straight stretches seen elsewhere throughout the middle section. Furthermore, at the "half turn" meander the narrow upland spur projecting into it shows clearly a slipoff slope at its south end and on its west (downstream) side. (b) Around the "half turn" meander, and upstream and downstream from it, for a total river distance of about 3-1/2 miles, the north wall of the canyon is very narrow and steep. (c) The "half-turn district" forms the only interruption to an otherwise continuous series of adjoining scallops (meander-migration scars) on the north side of the river. (d) The upland in the "half-turn district" includes a substantial outcrop of Weber sandstone—the only remnant of that formation north of the river in the middle section.

Taken together, these marked differences cannot plausibly be explained as due to coincidence. The "half-turn district" not only is exceptional in the middle section; it also shows conditions closely resembling those predominant in the lower section of the canyon, to which it is presumably related. Hence the lower section will next be described, before the "half turn district" is further discussed.


As herein designated, the lower section of the can you extends from point C (pl. 1), at the mouth of Big Joe Draw and of Starvation Valley, downstream for about 23-5/6 miles to point D, where the Yampa joins the Green east of Steamboat Rock.

The river surface has an altitude of 5,240 feet at point C and of 5,064 feet at point D. Thus in the lower section the river falls 176 feet, an average of nearly 7.4 feet per mile, which is less than half of the gradient of 16.9 feet per mile in the middle section.

As previously stated, the lower section differs markedly from the middle section in a number of ways, which will be discussed in detail in the pages that follow.


It will be recalled that in the middle section, except within the "half-turn district," Yampa River follows a course of open meanders interspersed with a few almost straight stretches. The average direction of that part of the river is N. 82° W., which is close to the regional strike of the rocks.

The lower section of the canyon differs notably from the middle section in its river pattern. First, near point C the river turns in a general southwesterly direction to the lower end of Bull Park (an airline distance of about 2-1/2 miles), before resuming its overall northwestward course to its junction with Green River at point D (pl. 1). Second, and more striking, the river's course is much more intricate and meandering; upland spurs alternate on the two sides of the river, and many of the meanders are so curved and interlocking as to be of the type called by Davis (1914, p. 23-24) "dove-tail."


In a general way the two walls of the canyon in the lower section resemble each other. But during their incision the meanders were not cut straight downward, for most cross sections of the canyon are asymmetric. Evidently lateral erosion and lateral movement of meanders have taken place, for in general the projecting ends and downstream sides of the spurs show slipoff slopes, whereas the upstream sides of the spurs and the walls on the outer side of meander curves show very steep slopes or even undercut and overhanging cliffs. (See figs. 4, 5.) Furthermore, meanders have become much more rounded during incision. However, the meander belt is still quite narrow and curving, the slipoff slopes occupy only parts of their respective spurs, and the bottom of the canyon is still very narrow and without conspicuous flood plain scrolls.

FIGURE 4.—"Tiger Wall," an overhanging cliff of Weber sandstone in lower section of Yampa Canyon. (National Park Service photograph.)

FIGURE 5.—Lower section of Yampa Canyon. Intricate meanders in Weber sandstone, view northward near Harding Hole. (National Park Service photograph.)

On a number of the interlocking spurs (such as the eight small spurs just downstream from Harding Hole and the several spurs just upstream from Warm Springs) the slipoff slopes are seen on the aerial photographs to be interrupted part way down by crude "treads" of somewhat less slope.

The so-called parks and holes (Bull, Harding, Burro, and Castle) in this section of the canyon are merely small, mostly steep floored, open spaces near the river, and are not comparable to Anderson, Tepee, and Browns Holes in the first, second, and third meander-migration scars of the middle section.

The depth of the canyon in the lower section, as measured from the river surface to the top of the cliff on the south side, differs through a wide range. As already described, the greatest depth observed is about 1,715 feet, at Warm Springs Cliff, 4 miles upstream from Green River. This cliff appears to expose the upper two-thirds of the Morgan formation, the full thickness of the Weber sandstone, and a few feet of the Park City formation that caps it. The least depth noted (excluding the small reentrants at the mouths of side streams) is 235 feet near the west end of Castle Park. Here, because of the southwestward dip of the beds and the large southward meander of the river, the south wall exposes only the upper part of the Weber sandstone just below its top.



The small area of upland from Schoonover Pasture across Johnson Canyon to East and West Serviceberry Draws is merely the western tip of the upland south of the middle section of the canyon, the topography of which has already been fully described.

The upland west and northwest of East and West Serviceberry Draws has its own characteristic topography. In general the upper beds of the Weber sandstone are exposed only in a narrow, irregular belt at the edge of the upland along the canyon; this belt is augmented here and there by exposures of the Weber up the side streams. Much of the upland is veneered by the overlying Park City formation, which at a number of places approaches, or even is the very top of, the canyon wall and extends southward and southwestward for distances up to a couple of miles. The Park City forms a resistant dip slope, the surface of which is rather smooth but, especially toward the west, is marked by a great number of very shallow channels and a few slightly larger ones draining southwestward down the dip. At or near the bottom of this dip slope these channels gather into larger channels extending northwestward or southeastward approximately along the boundary between the Park City and the next younger Moenkopi formation. In turn these larger channels empty into the few major streams (in Hells, Red Rock, and Sand Canyons) that succeed in flowing against the dip and joining Yampa River. (This topography and stream pattern are well shown on the topographic map and even better on the aerial photographs for the upland on the two sides of Sand Canyon.) It is interesting to note that between the combined Serviceberry Draw and point D, a river distance of nearly 20 miles, only 7 streams enter the south side of the Yampa—the 3 named above, and 4 others too small to be named on the topographic map.


Beginning near the river about half a mile downstream from point C, a high ridge extends northwestward to the east edge of the Warm Springs (sixth) scar, approximately parallel to and about half a mile southwest of Starvation Valley and the upper part of Warm Springs Draw. For most of its length the top of this ridge is higher than 7,000 feet; the highest point noted is marked "7365" on the topographic map. West of the Warm Springs scar the ridge resumes (though with a maximum altitude marked "6962") and extends westward for 1 mile to the edge of Lodore Canyon of Green River.

This ridge serves as a drainage divide, for it is not crossed by any of the streams that come from the high country still farther north. On the contrary, all those streams are deflected southeastward or northwestward along its northeastern base and together find a passage to Yampa River or the Green, or join Warm Springs and Iron Mine Draws, which extend down the Warm Springs scar.

The many streams that originate on the southwestern flank of the ridge extend southward and southwestward, down the dip, across a belt which, because of the river's sinuous course, ranges in width from 1/2 to 3 miles. The topography in this belt has been justifiably called by the Untermanns and others "fantastic." (See fig. 5.) Erosion in the poorly cemented and highly jointed Weber sandstone has produced a bewildering maze of sharp, narrow gorges. Most of these gorges begin at the ridge in deep, rounded or pointed amphitheaters; descend steeply, at places over hard ledges; and some finally drop abruptly to the river over dry "waterfalls" many feet in height.

A few patches of the thin overlying Park City formation are found at high spots on the ridge and in the belt south of it.

From its beginning near the river, northwestward for a distance of about 2 miles, the ridge is generally sharp crested and its northeastern flank is conspicuously crenulated. Farther northwest, the northeastern flank is much smoother.

The boundary between the Weber sandstone and the underlying Morgan formation follows very closely the base of the northeastern flank of the ridge and the adjoining southeastward- and northwestward-flowing streams described above.

Still farther away from Yampa River, in the Morgan and older formations, the topography is similar to that in the lower part of the Morgan and beds below it north of the middle section of the canyon, with many flatirons rising and pointing toward the north.


The most conspicuous and significant difference in geology between the middle and lower sections of Yampa Canyon is that, whereas in the middle section the boundary between the Weber sandstone and the underlying Morgan formation lies almost continuously high up along the south wall of the canyon, in the lower section that boundary lies predominantly at a substantial distance north of the river.

As already described, near point C Yampa River turns in an overall southwesterly direction to Bull Park, several miles away. This direction is down the dip, but as the gradient of the river is much less than the angle (6°) of dip, the amount of the Morgan formation exposed in the canyon dwindles rapidly downstream and ceases just below the mouth of Johnson Canyon. If this dwindling wedge of exposed beds of the Morgan is ignored, we may consider that near point C the Weber-Morgan boundary crosses from the south wall to the north side of Yampa River. Thence it extends northwestward up the floor of Starvation Valley and on to Lodore Canyon of Green River in a general course that is interrupted only at the Warm Springs scar, where the Morgan formation is again exposed southward to Yampa River.

Stated in a more summary way: the middle section of Yampa Canyon is eroded chiefly in the Morgan formation; the lower section, chiefly in the Weber sandstone.

Brief descriptions of the Weber and the Morgan, quoted from the Untermanns' report, are given on page I-9. It seems desirable to add here only the comment that on aerial photographs the jointing in the Weber sandstone, with a principal direction essentially that of the strike, is generally much more conspicuous in the lower section than in the middle and upper sections of the canyon.

As already indicated, a veneer of the thin Park City formation, lying above the Weber sandstone, holds up a fairly extensive dip slope south of the river and remains in a few patches north of the river in the lower section. According to the Untermanns (1954, p. 38), the Park City consists of "light gray to yellow, frequently silty or cherty, calcareous shale * * *. Gray thinly bedded cherty fossiliferous limestone and calcareous sandstone occur in the lower portion * * *. [In] the vicinity of Dinosaur National Monument headquarters, [the Park City] is only about 50 feet thick."


Just as the "half-turn district" is an exception to the topographic and geologic conditions that prevail in the middle section of Yampa Canyon, so the Warm Springs scar (see fig. 6) is an exception to the topographic and geologic conditions that predominate in the lower section.

FIGURE 6.—Warm Springs (sixth) scar on north side of lower section of Yampa Canyon. (Aerial photograph for Soil Conserv. Service, U.S. Dept. Agriculture, by Fairchild Aerial Surveys, Inc., 1938.)

The Warm Springs scar also is on the north side of the river. In size, shape, and degree of preservation, it somewhat more closely resembles the first and second scars than the other three. However, its southward-sloping floor is smoother, and the two streams that flow down it to the river—in Iron Mines and Warm Springs Draws—run in very shallow channels.

Like the five scars in the middle section, the Warm Springs scar is eroded in the Morgan formation, which it exposes southward to the river (and, still farther downdip, in the bottom of the canyon for several miles both upstream and downstream). Perhaps the most noteworthy difference between this scar and the other five is that both the east and west rims of the Warm Springs scar, in their southern half, include a substantial thickness of Weber sandstone above the Morgan formation.

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

Last Updated: 09-Nov-2009