Geological Survey Professional Paper 669 The Colorado River Region and John Wesley Powell

THE COLORADO RIVER REGION AND JOHN WESLEY POWELL

GEOLOGIC HISTORY OF THE COLORADO RIVER

By CHARLES B. HUNT

John Wesley Powell clearly recognized that the spectacular features of the Colorado River—its many grand canyons—were dependent upon the structural history of the mountainous barriers crossed by the river. He conceived of three different historical relationships between rivers and structural features: (1) Newly uplifted land surfaces have rivers that flow down the initial slope of the uplift; these relationships he termed consequent. (2) A river may be older than an uplift that it crosses because it has been able to maintain its course by eroding downward as the uplift progresses; this relationship he named antecedent. (3) An uplifted block may have been buried by younger deposits upon which a river becomes established. The river, in cutting downward, uncovers the uplifted block and becomes incised into it; this relationship he called superimposed.

The geologic history of the Colorado River involves all three relationships. In addition, although the position of the river course through a particular structural barrier may have been the result of superposition, the depth of the canyon at that point may be largely due to renewed uplift of the barrier; such deepening of the canyon, therefore, is due to antecedence. The problem of the Colorado River remains today very much as G. K. Gilbert stated it nearly 100 years ago: "How much is antecedent and how much is superimposed?" The question must be asked separately for each stretch of the river.

The geologic history of the Colorado River begins with the emergence of the Rocky Mountains and Colorado Plateau from the sea that had flooded them in Cretaceous time. During the early Tertiary (Paleocene and Eocene time, 65-40 million years ago), huge lakes in the northern part of the Colorado Plateau (Flagstaff and Greenriver Lakes) received consequent streams draining the west slope of the newly formed Rocky Mountains. Probably much of the southern part of the Colorado Plateau also drained north to the lakes, because the general dip of the Cretaceous and older rocks on the plateau is north, but the record of that assumed drainage has been lost because of erosion. The Oligocene and Miocene drainage history of much of the southern part of the Colorado Plateau is also obscure because datable deposits there are scarce. During Cretaceous time, the area that later became the Basin and Range province was higher than and drained into the area that later became the Colorado Plateau. This relationship probably continued during early Tertiary time.

By Oligocene time (40-25 million years ago), the lakes in the northern part of the Colorado Plateau had become filled with sediments, and the plateau was being tilted northeast. Filling of the lakes exceeded the rate of tilting, and they overflowed southward. It is assumed that the southward drainage again was ponded temporarily in the Henry Mountains and Kaiparowits basins, which lie south of the Uinta basin. No great amount of water need be involved in this supposed pending, because the rivers off the west slope of the Rocky Mountains were repeatedly beheaded by the lavas and by the basins and ranges forming there, and because the Green River was still trapped in the Wyoming basin.

Before the end of Oligocene time, the Gunnison River valley, oldest recognizable valley on the western slope of the Rocky Mountains, was eroded down to the Precambrian basement rock, and the eroded sediments were deposited in the lakes to the west. Later, the drainage became interrupted because the valley was tilted eastward and filled with lava and other eruptive materials.

The main stem of the Colorado River in the Rocky Mountains also began with a consequent course that drained westward to the early Tertiary lakes. That original consequent course was different from the present one. Gravel deposits (perhaps 25 million years old) under upper (?) Miocene lavas on the north flank of the White River Plateau show that the Colorado River originally flowed west from about Middle Park across what now is the headwaters of the Yampa River into the headwaters of the present White River that still follows an essentially consequent course down the trough of the Uinta structural basin.

During the Miocene (25-10 million years ago), the west-flowing consequent rivers became interrupted by block faulting that formed basins and ranges across the river courses. Uplift of the Gore Range checked the Colorado River for awhile in Middle Park, and renewed uplift of the White River Plateau further interrupted it west of the Gore Range. Fossiliferous sediments interbedded with the lava downfolded in the basins show that this interruption to the drainage occurred shortly after the middle of Miocene time. Lava in the Gunnison River valley, dated radiometrically, indicates that the drainage there also was interrupted, at least intermittently, during the Miocene.

About the end of Miocene time (about 10 million years go), the mountain barriers had been fully breached and the present drainage pattern established. The river courses across the structural barriers may have been superimposed in part, as the Gunnison River seems to have been where it crosses the uplift of Precambrian rocks at the Black Canyon. At other places, the river courses may have been started when the ponded rivers overflowed low places on the rim. At most of the structural barriers, however, uplift was renewed, and the canyons were deepened. Gore Canyon, Glenwood Canyon, Black Canyon of the Gunnison, for example, are in part antecedent.

The history of the Green and Yampa Rivers is similar. They became superimposed across the Uinta Mountains, probably when the mountains were partly buried by the Browns Park Formation of Miocene (?) age. At that time the mountains were lower relative to the adjoining basins than they are today. The Browns Park Formation in that area is much deformed, and the canyons of the rivers through the mountains probably owe much of their depth to late Tertiary and Quaternary uplift of the mountains (or downfolding of the basin); that is, the canyons are in part antecedent.

The San Juan basin probably began to overflow to the west in Oligocene time. Gravel deposits on the Kaibito Plateau south of Navajo Mountain show that the San Juan River had established its course westward across the Monument upwarp and was within 75 miles of the head of the Grand Canyon by late Miocene time. The gravels are regarded as older than earliest Pliocene (pre-Bidahochi Formation) and include pebbles of Miocene volcanic rocks from the San Juan Mountains.

Before middle Miocene time, a large canyon—as wide as the Grand Canyon and half as deep—had been eroded through the southwest rim of the Colorado Plateau. A segment of the canyon, preserved at Peach Springs, Ariz., is partly filled with deposits possibly related to the Muddy Creek Formation and dated radiometrically as 18.3 million years old. A gap at Kingman, Ariz., between the fault blocks of Precambrian rocks forming the Hualapai and Cerbat Mountains, 40 miles southwest of the present rim of the Colorado Plateau, is filled with lavas dated radiometrically as 16-17 million years old; the gap may be a segment of the Miocene canyon at Peach Springs that was faulted off the plateau at a later time. During Miocene and Pliocene time, a considerable area in central Arizona became separated from the Colorado Plateau by faulting and now is part of the relatively low-lying Basin and Range province.

We cannot be sure how much of the Colorado River basin drained off the plateau via the canyon at Peach Springs. Probably the Little Colorado River drainage was first to leave the plateau via that canyon. The Little Colorado River valley looks old. It is broad and open without deep narrow canyons (except at the mouth). Its original course seems to have been south of the Kaibab upwarp, and it may have been joined by the San Juan River. This ancestral course antedates the lavas in the San Francisco volcanic field and could be as old as Oligocene. By middle Miocene time, the canyon at Peach Springs was blocked by uplift and by deposits of volcanic materials and related sediments. Later, pounding of the San Juan and Little Colorado Rivers on the plateau east of the site of the present Grand Canyon seems to be recorded by playa or lake beds in the lower part of the Pliocene Bidahochi Formation.

The lower stretch of the Colorado River valley in the Basin and Range province was an estuary of the Gulf of California at various times during the Miocene and Pliocene. Pliocene marine shells have been found in the estuarine deposits as far north as Parker, Ariz., and very similar deposits without fossils are at The Needles, Ariz. These estuarine deposits and other deposits near the Miocene-Pliocene boundary in the Colorado River delta contain coccoliths and Foraminifera reworked from Cretaceous shale formations on the Colorado Plateau. Certainly there was through drainage from the Colorado Plateau by that time.

In the Lake Mead area, however, the earliest known Colorado River deposits seem to be no older than middle Pliocene and may be younger. During parts of Miocene time, when the river was depositing sediments from the Colorado Plateau in the estuary below The Needles, its course was probably via the canyon at Peach Springs and west from there perhaps via the gap at Kingman, Ariz.

The earliest deposit attributable to the Colorado River in the Lake Mend area is a curious limestone (Hualapai Limestone) that centers about the mouth of the lower Granite Gorge of the Grand Canyon. It was deposited in a lake 1,000 feet deep, yet there is no delta of clastic sediments at the mouth of the gorge. A major source of water was needed just to maintain the lake against evaporation, and the water had to be of a kind that would contribute much calcium carbonate without forming a delta. A possible explanation is that the Pliocene Colorado River, ponded in Grand Canyon above the dry canyon at Peach Springs, lost its water into the cavernous limestone, which is flexed there so that it provides a structural trough with more than 1,000-foot head, plunging from the dry canyon to the mouth of lower Granite Gorge. The first discharge of the Colorado River at the mouth of lower Granite Gorge may have been by big springs of the kind well known in limestone regions.

This lake, ponded in the structural basin immediately west of the Colorado Plateau, apparently overflowed a low place on the rim and cut the canyon at the Black Mountains. Tilted gravels indicate that the Black Mountains were further raised in late Pliocene and Quaternary time. The canyon is therefore partly antecedent.

At the west end of Lake Mead, the Colorado River turns 90° south to join its earlier course in the estuary at The Needles and farther south. This southerly stretch of the Colorado River separates two very different kinds of drainage systems. On the east is a structurally inactive block with the well-integrated Bill Williams and Gila River systems; on the west, the structurally active Mojave block, marked by abundant earthquake epicenters, recent fault scarps, and measurable present-day tilt, has no tributaries worth the name.

Quaternary erosion in the Colorado River basin seems to have been roughly proportional to the time involved, about 2 million years, about 3 percent of the Cenozoic. Throughout much of the Colorado Plateau and Rocky Mountains, about 500 feet of canyon deepening seems to have taken place during the Quaternary. There was greater erosion in the mountains at the valley heads, but radiometric dating of lavas in the bottom of the Grand Canyon suggests that the big canyon was within 50 feet of its present depth 1.2 million years ago.

The present sediment load of the Colorado River represents lowering of the river basin above the Grand Canyon at the rate of 6.5 inches per 1,000 years. Assuming this rate, and further assuming that there has been drainage off the southern part of the Colorado Plateau since middle Oligocene time and drainage from the Rocky Mountains and northern part of the plateau since late Miocene time, the Colorado Plateau and Rocky Mountains could have been lowered about 2 miles by the Colorado River system. This average seems to be about the right order of magnitude.