Geological Society of America Bulletin, published online on 26 January 2011 as doi:10.1130/B30274.1

The California River and its role in carving Grand Canyon

Brian Wernicke† Division of Geological and Planetary Sciences, California Institute of Technology, Pasadena, California 91125, USA

ABSTRACT the plateau. By Oligocene time, the lakes had basins may contain evidence of the time, place, largely dried up and were replaced by ergs. By and even rate of erosion, but impose few con- Recently published thermochronological mid-Miocene time, a pulse of unroofi ng had straints on the evolution of topographic form. and paleoelevation studies in the Grand Can- lowered the erosion level of eastern Grand A promising avenue of research in this other- yon region, combined with sedimentary prov- Canyon to within a few hundred meters of its wise discouraging endeavor stems from the fact enance data in both the coastal and interior present level, and the Arizona River drainage that isothermal surfaces in the uppermost crust portions of the North American Cordillera, below modern Grand Canyon was deranged more-or-less assume the geometry of ancient place important new constraints on the paleo- by extensional tectonism, cutting off the sup- topog raphy, leaving behind a sort of palimp- hydrological evolution of the southwestern ply of interior detritus to the coast. Increas- sest of the ancient landscape, especially in the United States. Review and synthesis of these ing moisture in the Rocky Mountains in late case of wide, deep canyons. After a period of data lead to an interpretation where incision Miocene time reinvigorated fl uviolacustrine kilometer-scale erosion, the most direct expres- of a large canyon from a plain of low elevation aggradation NE of the asymmetrical divide, sion of the ancient topographic form is its thermal and relief to a canyon of roughly the length which was fi nally overtopped between 6 imprint. By using thermal structure to reconstruct and depth of modern Grand Canyon occurred and 5 Ma, lowering base level in the interior ancient relief and comparing it to modern relief primarily in Campanian time (80–70 Ma). of the plateau by 1500 m. This event reinte- in the same mountainous region, fundamental Incision was accomplished by a main-stem, grated the former Arizona drainage system questions about landscape evolution may be ad- NE-fl owing antecedent river with headwaters through a cascade of spillover events through dressed. For example, does relief generally de- on the NE slope of the North American Cor- Basin and Range valleys, for the fi rst time crease as a function of time in a “geographical dillera in California, referred to herein after connecting sediment sources in Colorado cycle” of youth, maturity, and old age (Davis, its source region as the California River. At with the coast. This event, combined with the 1899), or does topographic form quickly attain a this time, the river had cut to within a few intensifi cation of summer rainfall as the Gulf “dynamic equilibrium” that changes little as ero- hundred meters of its modern erosion level of California opened, increased the sediment sion proceeds (Hack, 1960)? Although both end in western Grand Canyon, and to the level of yield through Grand Canyon by perhaps two members have been extensively discussed and ap- Lower Mesozoic strata in eastern Grand Can- orders of magnitude from its Miocene nadir, plied to the time and length scales of late Quater- yon. Subsequent collapse of the headwaters giving birth to the modern subcontinental- nary erosion (e.g., Heimsath et al., 1999), we are region into a continental borderland and co- scale Colorado River drainage system. The only just beginning to address whether ex trapo- eval uplift of the Cordilleran foreland during Colorado River has thus played a major role lation of these results applies to kilometer-scale the Laramide orogeny reversed the river’s in unroofi ng the interior of the Colorado Pla- erosion acting over time scales of 10–100 m.y. course by Paleogene time. After reversal, its teau, but was not an important factor in the (e.g., Reiners and Shuster, 2009). terminus lay near its former source regions in excavation of Grand Canyon. Relief on isothermal surfaces created by what is now the Western Transverse Ranges topography decreases exponentially downward and Salinian terrane. Its headwaters lay in the INTRODUCTION (e.g., Section 4–12 in Turcotte and Schubert, ancient Mojave/Mogollon Highlands region of 1982), so thermochronometric measurements at Arizona and eastern California, apparently How do landscapes evolve through signifi - depths within 1–2 times the amplitude of topog- reaching as far northeast as the eastern Grand cant amounts of geologic time? Because ero- raphy, or within the upper ~4 km of the crust Canyon region. This system is herein referred sion disaggregates rock masses (as opposed for most mountain belts, provide the best oppor- to after its source region as the Arizona River. to aggregating or modifying them), it presents tunity for reconstructing landscapes. Given that From Paleogene through late Miocene time, a special challenge for study. Most of what is temperatures at these depths are generally below the interior of the Colorado Plateau was a known about erosion concerns incremental 100 °C, the most effective thermochronometers closed basin separated from the Arizona River changes in modern landscapes. Unconformities for detecting this signal are fi ssion-track and drainage by an asymmetrical divide in the provide valuable records of the form of ancient (U-Th)/He dating of apatite (e.g., Stüwe et al., Lees Ferry–Glen Canyon area, with a steep erosion surfaces, but only provide a snapshot 1994; House et al., 1998, 2001). SW fl ank and gently sloping NE fl ank that of the transition from erosion to aggradation. Debate over the origin of Grand Canyon, the drained into large interior lakes, fed primar- When regionally developed, they are generally planet’s most vivid illustration of kilometer- ily by Cor dilleran/Rocky Mountain sources to cut on surfaces of very low relief. Kilometer- scale erosion, has been invigorated over the the north and west, and by recycled California scale topo graphic forms characteristic of moun- last two years by application of these and other River detritus shed from Laramide uplifts on tain belts, if preserved at all by unconformities, proxies for erosion and paleoelevation in the cover only a small fraction of eroding uplands. region (e.g., Flowers et al., 2008; Hill et al., †E-mail: [email protected] Studies of the eroded detritus in sedimentary 2008; Hill and Ranney, 2008; Karlstrom et al.,

GSA Bulletin; Month/Month 2010; v. 1xx; no. X/X; p. 1–29; doi: 10.1130/B30274.1; 14 fi gures.

For permission to copy, contact [email protected] 1 © 2011 Geological Society of America Geological Society of America Bulletin, published online on 26 January 2011 as doi:10.1130/B30274.1

Wernicke

2008; Pearthree et al., 2008; Pederson et al., spired the classical concepts of antecedent and younger, developed on post-tectonic fi ll (e.g., 2008; Polyak et al., 2008a, 2008b; Young, superposed drainage (Powell, 1875). In ante- Emmons, 1897). 2008). Grand Canyon is a long, relatively wide cedence, the erosive power of a stream formed Over more recent decades, a contrary consen- canyon through which surface waters of a large in a region of low relief is suffi cient to maintain sus has emerged, holding that incision of Grand area of the southwestern U.S. interior pass be- its grade during tectonic distortion of the land- Canyon began in late Miocene time, when two fore ultimately reaching the Gulf of California. scape, whereas in superposition, the stream previously separate drainage basins became The high plateaus surrounding Grand Canyon originates in fl at-lying post-tectonic strata, and inte grated (e.g., Longwell, 1946; McKee et al., constitute the most imposing of a series of cuts downward across structure. The fi rst few 1967; Lucchitta, 1972, 2003). There is general kilometer-scale topographic obstacles (e.g., the decades of exploration of these canyons thus consensus that integration occurred between 6 Kaibab arch, Fig. 1) that the Colorado River and led to a general debate about whether these and 5 Ma, before which an older upper basin its primary northern tributary, the Green River, rivers were older than the Laramide structures and a younger lower basin were separated by a cut improbably across as “transverse drain- they cut through and therefore antecedent (e.g., drainage divide somewhere in the vicinity of the ages” (e.g., Douglass et al., 2009), which in- Powell, 1875; Dutton, 1882; Walcott, 1890), or Kaibab arch (e.g., Hill et al., 2008; Karlstrom

113°W 111°W

NV UT Kaibab escarpment Kaiparowits Plateau E limit, major T extension

B a s i n Plateau boundary

a r c h UT Bidahochi Fm. (Miocene) Rim gravels, Claron Fm. Lees AZ (Paleogene)

b Grand Upper K strata

a Ferry

Wash Mtns. b Trough i

a BASAL UNCONFORMITIES FOR:

C A N Y O K N Cretaceous C o c o n i n o Triassic Virgin Cambrian UG

36°N 36°N r K e LP G R A N D

v LG

i T e r r a c e R Figure 2 Tr-J NV a n d CA

Plz o T r a n s i t i o n Z o n e

Arizona homocline d a

pC r R a n g e o

l 34°N 34°N o

C 100 km Ft. Apache 113°W 111°W

Figure 1. Tectonic map showing selected geographical and geological features discussed in text. Geology of Colorado Plateau and Transition Zone is generalized with late Cenozoic volcanic units removed. Boundary of Coconino terrace (light-blue area), based on regional elevation of the top of the Kaibab Formation at 1600 ± 200 m, is based on contour map of Hunt (1969). LG—Lower Granite Gorge of Grand Canyon; LP—Long Point area; UG—Upper Granite Gorge; p C—Proterozoic crystalline and overlying Proterozoic stratifi ed rocks; Plz—Paleozoic strata; Tr-J—Triassic and Jurassic strata; K—Cretaceous strata.

2 Geological Society of America Bulletin, Month/Month 2010 Geological Society of America Bulletin, published online on 26 January 2011 as doi:10.1130/B30274.1

California River et al., 2008; Pearthree et al., 2008; Pederson and 200 km across, with the structural terrace normal fault blocks and associated basement- et al., 2008). Integration would thus require occupying a 200-km-long, 50–100-km-wide cored uplifts that were tilted during a middle either piracy of the upper basin by the lower, or area within its NW portion (Hunt, 1969). The Miocene pulse of extensional deformation lateral spillover from the upper basin into the structural terrace lies at a mean elevation of (Wernicke and Axen, 1988; Brady et al., 2000; lower, or perhaps some combination of the two 1900 m, and its erosion surface is almost exclu- Billingsley et al., 2004; Wallace et al., 2005; (e.g., Hunt, 1969; Spencer and Pearthree, 2001; sively cut near the contact between the Permian Quigley et al., 2010). Pederson, 2008). Following in the footsteps of a Kaibab and Triassic Moenkopi Formations, or In Grand Wash Trough and other basins to long history of debate about the evolution of pre– on unconformably overlying Tertiary basalts. the west, the tilted fault blocks are primarily Grand Canyon drainages (reviewed in Powell, The principal structural features disrupting this overlain by >500 m of gently tilted to fl at-lying 2005), various and contrasting proposals have pattern are relatively modest net offsets along Miocene basin fi ll, deposited between 13 and recently been made for at least some pre–6 Ma the N-trending Hurricane and Toroweap faults 6 Ma, generally referred to as the Muddy Creek incision of signifi cant portions of Grand Can- in the western part of the canyon, and the Kai- Formation (Fig. 3; Longwell, 1936; Lucchitta, yon (e.g., Polyak et al., 2008a; Hill and Ranney, bab arch along the eastern part (e.g., Karlstrom 1979; Wenrich et al., 1996; Karlstrom et al., 2008), both upstream and downstream of the et al., 2008), both of which are associated with 2008) or the informal designations “rocks of the Kaibab arch, with varying amounts of incision innumerable smaller monoclinal fl exures and Grand Wash Trough” (Bohannon, 1984; Wal- dating perhaps as far back as Laramide time faults (e.g., Billingsley et al., 1996). Within the lace et al., 2005) and “sedimentary rocks of the (Late Cretaceous through Eocene; e.g., Flowers Kaibab arch, the canyon rim gradually rises Grand Wash Trough” (Billingsley et al., 2004). et al., 2008; Young, 2008). Even these studies, eastward an additional 400 m before descend- Within the trough, the modern Colorado River however, did not view pre–6 Ma incision to have ing abruptly at the canyon’s eastern terminus. drainage system dissects a 6 Ma fi ll surface and been accomplished by an antecedent river cross- Because of the importance of the homocline underlying rocks to a depth of ~500 m, exposing ing the arch, nor did they challenge the piracy/ and structural terrace to the discussion, I will the three-dimensional geometry of the Muddy spillover paradigm. refer to them informally herein as the “Ari- Creek depositional basin and its pre-Tertiary In this paper, I review the geologic setting zona homocline” and “Coconino terrace,” with substrate (Fig. 3). Near the modern level of Lake and contemporary thinking about the origin the understanding that the former includes Mead at Sandy Point (Fig. 2), thin deposits of of Grand Canyon, which revolves around an parts of southern Utah and the latter is more well-rounded fl uvial gravel of similar compo- issue informally known among analysts as “the extensive than the physiographic Coconino sition to the modern Colorado River bed load Muddy Creek problem.” I then review critical Plateau (Fig. 1). unconformably overlie the Muddy Creek For- evidence bearing on (1) the thermal history of In Grand Canyon, the Colorado River cuts mation and are intercalated with a basalt fl ow the shallow crust in the region, (2) the Cenozoic downward through the Paleozoic section and dated at 4.4 Ma (Howard and Bohannon, 2001). elevation and climatic history of the southern into basement rocks between Lees Ferry and the Upper Mesozoic and Paleogene strata along portion of the Colorado Plateau, and (3) the Upper Granite Gorge, and then runs at a level the NE fl ank of the Arizona homocline crop relationship between the timing and geometry near the basal Cambrian unconformity. It de- out mostly in southern Utah and northeastern- of Grand Canyon erosion and the provenance of scends at a nearly constant gradient of ~1.3 m/km most Arizona (Fig. 1). The SW portion of the depocenters (a) within the Grand Canyon region through the canyon, from an elevation of 940 m Arizona homocline lies within the Arizona (Rim gravels and other deposits), (b) upstream at Lees Ferry near its eastern end, to 360 m at its Transition Zone, a NW-trending, 500 km × of Grand Canyon in Utah, and (c) downstream western end where it enters Grand Wash Trough 100 km region that includes mainly Protero- of Grand Canyon in southern California. These (Fig. 1). There, the river crosses the abrupt tran- zoic basement exposures, exhibiting a structural constraints are synthesized into a new fi rst- sition between the Colorado Plateau and Basin and physiographic gradation between the little- order paleohydrological reconstruction of the and Range physiographic provinces, and en- faulted Colorado Plateau at 1900 m elevation region that now includes the southern portion ters the topographically lower surroundings of and the highly extended southern Basin and of the Colorado River drainage basin from Grand Wash Trough (Fig. 2). The westernmost Range Province, generally at <500 m elevation Campanian to Quaternary time, reconciling it segment of the canyon trends NW along the SW (Fig. 1; e.g., Elston and Young, 1991; Foster with the Muddy Creek problem and new con- margin of the Coconino terrace (Fig. 1). Imme- et al., 1993; Potochnik, 2001). Throughout the straints on the late Quaternary incision rate of diately SW of this river segment, the Paleozoic Transition Zone and much of the Basin and Grand Canyon. section acquires a gentle NE dip and forms the Range, Tertiary strata lie nonconformably on Hualapai Plateau (Fig. 2). basement rocks. Astride the boundary between Geologic Setting Grand Wash Trough is bounded on its east the Colorado Plateau and the Transition Zone, side by a rampart of fl at-lying Paleozoic strata Paleogene deposits, known informally as the Grand Canyon is a sinuous, 300-km-long, known as the Grand Wash Cliffs, and on its west “Rim gravels” (Fig. 1), indicate NE paleofl ow 15–20-km-wide, and ~1500-m-deep gorge side by the Virgin Mountains (Fig. 2). Near the (Elston and Young, 1991; Potochnik, 2001). through the southwestern Colorado Plateau center of the trough, a low, narrow ridge named Mid-Tertiary and younger gravels in the Transi- (Fig. 1). Although not the deepest, because of its Wheeler Ridge consists of steeply E-tilted Paleo- tion Zone indicate NE to SW fl ow similar to the extraordinary length, it is volumetrically among zoic bedrock unconformably overlain by vari- modern regime, indicating that a reversal in the the largest river gorges on Earth. It is cut mainly ably E-tilted Tertiary strata, ranging from 24 to drainage pattern occurred in mid-Tertiary time within a structural terrace (fl at-lying strata be- 18 Ma deposits of the Rainbow Gardens Mem- (e.g., Peirce, 1979; Holm, 2001). Prior to the re ver- tween two monoclines) that interrupts an other- ber of the Horse Spring Formation (Bohannon, sal, erosional unroofi ng of the Transition Zone wise gently NE-dipping (0.4°) homocline of 1984; Beard, 1993) to gently tilted 4.4 Ma Colo- and areas to the SW is regarded by most work- cratonic Paleozoic through Tertiary strata that rado River gravels (Fig. 3; Howard and Bohan- ers as focused on the ancient “Mogollon High- forms the SW quadrant of the plateau (Fig. 1). non, 2001; Howard et al., 2008). Wheeler Ridge lands” of central Arizona or “Kingman arch” The homocline measures ~500 km along strike is the easternmost of a series of domino-like of northwestern Arizona, fl anked by a drainage

Geological Society of America Bulletin, Month/Month 2010 3 Geological Society of America Bulletin, published online on 26 January 2011 as doi:10.1130/B30274.1

Wernicke

Azure Ridge

M t n s.

T r o u g h S h i v w i t s

C l i f f s Sandy Line of section, Point R i d g e Figure 3

V i r g i n Line of section, W h e e l e r W a s h Figure 12B P l a t e a u

S . Figure 2. Shaded relief map of western Grand Canyon–Lower G r a n d Granite Gorge region showing 36°N locations of sections and map in Figures 3, 8, and 12, the loca- Map area, Line of section, tion of paleocanyons (general- Figure 12A Figure 8 ized from Young, 2008), sample G r a n d W a s h locations of the three (U-Th)/He n apatite ages shown in Figure 5 H u a l a p a i 71±3 Ma 89±7 Ma 75±10 Ma and discussed in text (magenta o circles; after Flowers et al., 2008), and selected geographi- C y a Diamond cal features discussed in text. n Creek Hindu Can. ash W

ed Can. e w

Rim gravel d n a r G Springs Milk Plateau paleocanyons

Peach

Music Mtn. 01020 2035 m Peach N km Truxton Springs

114°W

network that carried detritus NE onto lowlands Grand Canyon and Fort Apache regions (e.g., The Muddy Creek Problem that now form the Colorado Plateau (e.g., Elston Young, 2001a). According to a reconstruction and Young, 1991; Potochnik and Faulds, 1998; of a regional cross section through these widely Debate over the origin of Grand Canyon Potochnik, 2001; Dickinson and Gehrels, 2008). exposed unconformities by Flowers et al. (2008; in particular, and therefore the paleohydrol- The rise of the Mogollon Highlands was see also Potochnik, 2001), of a total 5500 m of ogy and paleotopography of the southwestern protracted, as suggested by two regional un- structural relief on the basal Cambrian uncon- United States in general, has intensifi ed over conformities within the Arizona homocline, one formity, ~2300 m had developed by ca. 94 Ma the last three years, owing to the publication of overlain by Cretaceous strata and the other by the (late Cenomanian), and an additional 3200 m a wealth of new or greatly refi ned geochrono- Paleogene Rim gravels, both of which progres- developed between ca. 80 and 50 Ma. These data logical and thermometric measurements bear- sively cut downsection toward the SW. The sub- indicate that a signifi cant fraction of the struc- ing on the canyon’s history. These include Cretaceous unconformity cuts from a substrate tural relief across the Arizona homocline is pre- (1) U/Pb dating of speleothems in and near the of Upper Jurassic strata in northeastern Arizona Laramide (e.g., Potochnik, 2001). Whereas the canyon walls (Polyak et al., 2008b); (2) high- down to Permian strata along the SW edge of the older unconformity is cut on a surface of perhaps precision 40Ar/39Ar dating of basalts inter- plateau in the Fort Apache region of the Tran- a few tens of meters of local relief, the sub–Rim calated with river-terrace deposits (Karlstrom sition Zone (Fig. 1; e.g., Finnell, 1966; Hunt, gravel unconformity preserves local relief of at et al., 2007; Crow et al., 2008); (3) (U-Th)/He 1969). The sub–Rim gravel unconformity cuts least 1200 m on the Hualapai Plateau, just south dating of igneous and detrital apatites across the abruptly downward from Permian-Triassic strata of the westernmost segment of Grand Canyon region (Flowers et al., 2008); (4) U/Pb dating of on the SW margin of the plateau to Protero zoic (e.g., Young, 1979), and at least 850 m in the Fort large populations of detrital zircons in sedimen- basement in the Transition Zone, in both the Apache region (Potochnik, 2001). tary basins related to regional erosion (Larsen,

4 Geological Society of America Bulletin, Month/Month 2010 Geological Society of America Bulletin, published online on 26 January 2011 as doi:10.1130/B30274.1

California River

Shivwits WNW ESE BREAK Plateau 2000 Grand Wash Trough Grand Wash Cliffs 2000

Buttress Sanup Plateau ca. 6 Ma fill surface unconformity PP 1600 m Hualapai Limestone 1000 Sandy Point M 1000 Tm CD RIVER/LAKE LEVEL

ELEVATION (m) ELEVATION Tm 0 0 Plz Xg Xg Th

WHEELER GRAND WASH FAULT Wheeler Ridge (buried) FAULT 10 km VERTICAL EXAGGERATION 4:1

Figure 3. Cross section showing stratigraphic and structural relations at the western terminus of Grand Canyon. Section to the left of the break is modifi ed after Wallace et al. (2005), and section to the right of the break is modi- fi ed after Billingsley et al. (2004). Xg—Proterozoic basement; CD—Cambrian through Devonian strata; M—Mis- sissippian strata; PP—Pennsylvanian and Permian strata; Plz—Paleozoic strata undifferentiated; Th—Horse Spring Formation; Tm—Muddy Creek Formation (sedimentary rocks of Grand Wash Trough). See Figure 2 for location of section.

2007; Link et al., 2007; Dickinson and Gehrels, a 7 m.y. depositional span from 13 to 6 Ma Polyak et al., 2008b; Hill and Ranney, 2008; 2008, 2009; Jacobson et al., 2010; Jinnah across the western terminus of Grand Canyon, Young, 2008). As noted already, young incision et al., 2009; Davis et al., 2010; Larsen et al., the Muddy Creek near the mouth of Grand necessitates connection of the upper and lower 2010); and (5) paleo altimetry of the plateau Canyon does not exhibit sedimentary struc- parts of the basin through piracy or spillover using “clumped isotope” thermometry of lacus- tures indicative of fl uvial deposi tion or a large (McKee et al., 1967). Although the element of trine carbonates (Huntington et al., 2010). The river system emanating from Grand Canyon connecting the two drainages across the Kaibab torrent of fresh information has renewed com- (Longwell, 1946; Lucchitta , 1966, 1979; Hunt, arch and Coconino terrace sensibly resolves the munity interest in what has long been the cen- 1969; Faulds et al., 2001; Wallace et al., 2005). Muddy Creek problem, it also leads to a cascade tral issue in explaining the history of the canyon, The core of the Muddy Creek problem is, thus, of puzzling consequences that have fueled con- namely the “Muddy Creek problem” (Black- reconciling the relatively slow accumulation troversy over the past 70 years. welder, 1934; Longwell, 1936, 1946; Lucchitta, of a small volume of locally derived sediment Leaving Grand Canyon substantially un- 1972; Faulds et al., 2001; Pederson, 2008). in a Miocene half-graben with a hypotheti- carved at 6 Ma requires no ordinary integra- As concluded by Karlstrom et al. (2008, cal pre–6 Ma Colo rado River that would have tion, but integration of a highly organized, p. 835), fl owed through it during this time. As summa- focused extant upper drainage through the edi- rized by Karlstrom et al. (2008, p. 835), fi ce of the westernmost plateau (McKee et al., After over a century of controversy there is a growing 1967; Lucchitta , 2003). The ensuing problems consensus that Grand Canyon has formed in the past Evidence for inception of carving of the Grand Can- 6 Ma (Young and Spamer, 2001). In this consensus, yon after 6 Ma is strong. (1) The sedimentary record include: (1) the location of the abandoned the term Grand Canyon is used for the canyon system shows that there are no Colorado River sediments in downstream reach of the upper basin prior to carved by a west-fl owing Colorado River, not for lo- the 13–6 Ma Muddy Creek Formation that now blan- inte gra tion (suggested courses have included the cal precursor canyons (Young, 2008), or for northeast- kets the Grand Wash Trough at the mouth of Grand Rio Grande according to McKee et al. [1967]; fl owing Tertiary drainages that may have existed in Canyon (Lucchitta, 1972; Faulds et al., 2001). (2) The western Arizona according to Hunt [1969]; now-eroded Mesozoic strata (Flowers et al., 2008). fi rst sediments containing distinctive sand composi- tion and detrital zircons that can be traced to Rocky southern Utah via the southern Kaibab arch, ac- The cornerstone of this consensus is that Mountain sources reached the newly opened Gulf of cording to Lucchitta [1984]); (2) the reason why any hypothesis favoring the existence of a California at 5.3 Ma (Dorsey et al., 2007; Kimbrough a well-integrated, subcontinental-scale drainage pre–6 Ma Colorado River cannot account for et al., 2007). (3) Gravels on top of the 6 Ma Hualapai would have changed its course in the middle of Limestone and beneath the 4.4 Ma Sandy Point basalt the fact that the Muddy Creek Formation in show that the river became established in its present a tectonically stable block; and (3) the require- Grand Wash Trough, which includes coeval course between 6 and 4.4 Ma. (italics added) ments of (a) headward erosion from the Grand deposits assigned to the Hualapai Limestone Wash Cliffs, (b) spillover of some form of up- (Fig. 3), contains exclusively locally derived At issue is whether these facts preclude stream dam >150 km east of the Grand Wash detritus in lacustrine and alluvial-fan depo- signifi cant—or nearly complete—carving of Cliffs at 6 Ma, or (c) both. sitional facies, contrasting strongly with the Grand Canyon prior to 6 Ma. The observations Headward erosion from the Grand Wash rounded detritus of exotic provenance char- behind the consensus relate to drainage integra- Cliffs raises the question of (4) why one of a acteristic of the bed load in the modern river tion, and do not contain direct information about series of small, arid canyons without perennial channel. Despite continuous exposures and when incision of Grand Canyon occurred (e.g., streams, similar to adjacent canyons now cut into

Geological Society of America Bulletin, Month/Month 2010 5 Geological Society of America Bulletin, published online on 26 January 2011 as doi:10.1130/B30274.1

Wernicke the cliffs, would spontaneously develop into one tions of fi ssion tracks that cluster at ~14.0– tectonism (e.g., Fitzgerald et al., 1991, 2009; of the great erosional spectacles of the planet 14.5 µm long (e.g., Fitzgerald et al., 1991). For Reiners et al., 2000). This connection is critical (e.g., Spencer and Pearthree, 2001). In the apatite grains that reside below 110 °C for long because it allows comparison of the pre–17 Ma gentle sarcasm of Hunt’s (1968) book review of periods of time, partial annealing of the tracks cooling history of the canyon with that of the McKee et al. (1967), it would have been a “pre- occurs, which may result in a younger age than surrounding plateaus. In the next section, fol- cocious gully” indeed. In regard to an upstream the time at which the samples cooled through lowing convention, “apatite fi ssion-track” will dam, evidence for some form of pre–6 Ma 110 °C. Depending on the thermal history, be abbreviated as AFT, “apatite helium” as ponding immediately upstream of Grand Can- populations of track lengths exhibit means that AHE, “partial-annealing zone” as PAZ, and yon is good—the fl uviolacustrine Bidahochi are typically 10%–20% shorter than those of “partial-retention zone” as PRZ. Formation blanketed a large area of northeast- rapidly cooled samples and exhibit a greater ern Arizona between 16 and 6 Ma, and slow, variance about the mean (e.g., Kelley et al., Cooling Histories of Samples in the Eastern episodic aggradation appears to have given way 2001). Although 60 °C has been regarded as Grand Canyon Region to rapid erosion at ca. 6 Ma, perhaps as a result a nominal lower limit for signifi cant anneal- of the integration recorded downstream (e.g., ing, track length reductions of up to 11% have Apatite Fission-Track Ages Scarborough, 2001; Meek and Douglass, 2001; been observed in natural apatites residing well AFT ages from basement rocks in the east- Douglass et al., 2009). The existence of other, below 60 °C over geologic time scales (e.g., ern, deepest segment of the canyon near the now-eroded Miocene ponding sites in the Glen Spiegel et al., 2007). crest of the Kaibab arch are 63 ± 2 Ma, with Canyon area is certainly possible (Hunt, 1969; For (U-Th)/He analysis, at geologic time track lengths of ~12.0–12.6 µm, indicating Hill and Ranney, 2008). However, as long noted scales, apatites that have not been damaged by signifi cant residence in the PAZ after 65 Ma by many (e.g., McKee et al., 1967; Hunt, 1969; radioactivity do not retain helium above 70 °C, (Dumitru et al., 1994). The data of Naeser et al. Hill et al., 2008), (5) the rim of Grand Canyon and retain most or all of it below 30 °C (e.g., (1989) and Kelley et al. (2001), which cover a at the point where spillover would have occured Farley, 2000). Partial retention occurs between broader area of the Upper Granite Gorge, show along the crest of the Kaibab arch, at roughly 30 °C and 70 °C. For samples with signifi cant a scattering of ages between 70 and 30 Ma, 2250 m elevation, lies ~300 m above the residence time below 110 °C, age may correlate with mean track lengths as low as 10.7 µm. The Bidahochi Formation, and towers 1300 m above strongly with the effective U concentration (eU) short track lengths require signifi cant residence the current elevation of the river where it enters in the sample. Higher eU samples are much time in the PAZ, and as such, the measured Grand Canyon south of Lees Ferry (Fig. 1). Any more retentive of He than lower eU samples, ages generally underestimate the age at which such lake, even if it were as deep and areally ex- because radiation damage creates traps that are the samples passed through 110 °C. Assuming tensive as would be needed, seems more likely retentive of He (Shuster et al., 2006; Flowers a proportionate scaling between age reduction to have drained into much lower ground that et al., 2007). For igneous apatites, variation in and track-length reduction of 1:1 for the oldest, exists to the north and south of the modern can- eU within samples may be quite limited. None- least annealed samples (Green, 1988), Dumitru yon in avoidance of the structurally high crest theless, samples from neighboring plutons may et al. (1994) estimated an age of cooling through of the Kaibab arch (e.g., Spencer et al., 2008a; exhibit a wide variation in eU and age. For detri- 110 °C of 75 ± 6 Ma for samples at the bottom Douglass et al., 2009). tal apatites, a single sample may yield a popula- of the Upper Granite Gorge in the vicinity of the The enigmatic consequences of piracy or tion with wide variation in eU and age. In both Kaibab arch (point B, Fig. 4A). spillover across the Kaibab arch have led to plutonic and detrital samples, the variation of Forward modeling of age and track-length necessarily elaborate hypotheses involving cut- age as a function of eU in some instances can be distributions using a Monte Carlo approach ting of most of eastern Grand Canyon during modeled to constrain the thermal history of the (e.g., Ketcham et al., 1999) suggests two pulses the Laramide to mitigate the height issue of the area (Flowers et al., 2007, 2008). of cooling in this area, the fi rst through 110 °C spillover point (Scarborough, 2001), or inte- Existing apatite fi ssion-track and He data between 80 and 70 Ma, residence at tempera- gration via a period of subterranean underfl ow for Grand Canyon and environs include tures of 60 ± 10 °C through most of Tertiary through the regional Cambrian-Mississippian fi ssion-track analyses of both igneous samples time, followed by fi nal cooling to surface tem- carbonate aquifer, at fi rst feeding the Hualapai from the basement and detrital samples from peratures in the late Tertiary (curve B, Fig. 4C; Limestone at the foot of the Grand Wash Cliffs, Protero zoic and Phanerozoic strata. Extant Kelley et al., 2001). and ultimately developing karstic collapse along data are suffi cient to infer cooling histories of The modeling by Kelley et al. (2001) also the modern course of the river (e.g., Hunt, 1969; samples within eastern and western Grand Can- suggests differences in both ages and cooling Pederson, 2008; Hill et al., 2008). yon (Upper Granite Gorge and Lower Granite histories across a small fault in the Upper Gran- Gorge areas, respectively, Fig. 1), as well as ite Gorge (~100 m net offset), with tempera- THERMAL HISTORY OF THE in tilted fault blocks in the Virgin Mountains tures ~10 °C warmer on one side of fault than SHALLOW CRUST IN THE (Azure Ridge area, Fig. 2) that had a full Paleo- on the other through most of Tertiary time, in GRAND CANYON REGION zoic section on top of them until ca. 17 Ma turn suggesting 200–300 m of offset of isother- (Fig. 4). Although now within the Basin and mal surfaces. Alternatively, because similar age Methods Range tectonic province, prior to ca. 17 Ma, variations also occur over distances of as little the Virgin Mountains resided in the cratonic as 10 km where no structures exist, they may At cooling rates typical of erosional unroof- foreland of the retroarc Cordilleran fold-and- simply be a function of contrasting annealing ing, common primary igneous and detrital thrust belt (Sevier orogen; e.g., DeCelles , kinetics in different rock types, which in any apatites completely anneal fi ssion tracks above 2004), in stratigraphic and structural continuity event are poorly understood for samples near 110 °C (e.g., Laslett et al., 1987). Samples that with the modern Grand Wash Cliffs. Therefore, the top of the PAZ (~60 °C). For example, the cool quickly from temperatures above 110 °C its thermal history is a reasonable proxy for South Virgin Mountains contain one of the most to near-surface temperatures preserve popula- that of the adjacent plateau prior to extensional extensively studied fossil PAZs in the world.

6 Geological Society of America Bulletin, Month/Month 2010 Geological Society of America Bulletin, published online on 26 January 2011 as doi:10.1130/B30274.1

California River

A Eastern Grand Canyon–Upper Granite Gorge 20 km Kaibab E. Grand Coconino NORTH Plateau Canyon Plateau SOUTH 5 80 Ma 5

4 4

3 A 3 70 Ma

4 Figure 4. Schematic cross sections show- AHE 19 2 2 Plz 20 Ma 4 ing representative thermochronological AHE 21 AHE 22 4 B AFT 120 5, 6 Plz data, low-temperature thermal histories, 1 1 Present elevation (km) elevation Present Xg and erosion histories for Grand Canyon re- AHE 23 4 Xg 0 0 gion. Erosion histories assume a geothermal AFT 75 (12µm) 5, 6 gradient of 25 °C/km. (A) Section showing positions of rim and gorge samples (circles) and inferred position of the erosion surface B Western Grand Canyon–Lower Granite Gorge (colored lines) at the times indicated on the N. Virgin S. Virgin Mtns.– W. Grand Hualapai Music SOUTH NORTH curves in the eastern Grand Canyon re- Mountains W. Coconino terrace Canyon Plateau Mtn. 4 4 gion. AHE—(U-Th)/He apatite age; AFT— 80 Ma fi ssion-track apatite age, showing mean 3 3 track lengths in parentheses. Here, repre- AHE 68 4 sentative ages shown are not averages of 70 Ma 2 2 measured ages, but calculated times of rapid Mz Plz Plz cooling through most or all of the AFT par- 1 D 1 tial annealing zone or AHE partial retention C Xg zone, as discussed in text. (B) As in A for Plz 7 4 0

Present elevation (km) elevation Present 0 AHE (15) AHE 71 3 5 western Grand Canyon; ages for point D are AFT 50 (12µm) AFT 75 (14µm) Peach Springs-Truxton from the Diamond Creek area, where both –1 Xg paleocanyon –1 2 AHE and AFT data exist. Sources of data: AFT 15 (14µm)1 AHE 15 3 1—Quigley et al. (2010); 2—Reiners et al. AFT 15 (14µm) (2000); 3—Fitzgerald et al. (1991, 2009); C Thermal histories 4—Flowers et al. (2008); 5—Kelley et al. Incision of Grand Canyon (2001); 6—Dumitru et al. (1994); 7—age in- 150 ferred from structural position between the 150 surface and the base of the partial annealing Erosional unroofing of zone for apatite fi ssion tracks (Fitzgerald E. Grand Canyon region B Δ Δ et al., 1991; Reiners et al., 2000). (C) Cooling T – TAB ~ 40 °C histories for the four lettered sample posi- T (°C) 100 C 100 tions in A and B, and discussed in text. Tectonic unroofing of Virgin Mtns. D A Δ Δ 50 50 TD – TC ~ 40 °C

0 50 100 Time before present (Ma)

There, the ages of samples near the top of the the topographic relief, and apparent variations (Dumitru et al., 1994) or below (Kelley et al., PAZ exhibit ~30 m.y. of variation at any given in Tertiary residence temperatures represent at 2001) the basement samples. From 60 to 20 Ma, depth, similar to the pattern in the Upper Granite most only 25% of the total amount of cooling the base of the PAZ at 110 °C would had to have Gorge (fi g. 5 in Fitzgerald et al., 2009). Hence, that occurred from 80 to 70 Ma. been at least 1500 m deeper in order to maintain although some of the variation in AFT age AFT ages in samples above the basement the samples near 60 °C, assuming a maximum may be the result of minor faulting, overall the generally increase rapidly upward within the geothermal gradient of 30 °C/km. This provides structural relief on the basal Cambrian uncon- Phanerozoic section, suggesting that the base a minimum estimate of the amount of Late Cre- formity in the inner gorge is a small fraction of of the PAZ (110° isotherm) at 80 Ma was near taceous (Campanian–Maastrichtian) erosion of

Geological Society of America Bulletin, Month/Month 2010 7 Geological Society of America Bulletin, published online on 26 January 2011 as doi:10.1130/B30274.1

Wernicke eastern Grand Canyon of ~1500 m, similar to Oligocene or early Miocene time, which fairly section in this fault block (Brady et al., 2000) the minimum estimate of 1200 m suggested by evenly unroofed both gorge and rim down to indicates a pretilt depth range for the 110 °C Dumitru et al. (1994; Figs. 4A and 4C). erosion levels not far above the modern topog- isotherm of 3500–4100 m. Assuming a surface raphy (Flowers et al., 2008). elevation near 2000 m at 17 Ma (Huntington Apatite (U-Th)/He Ages et al., 2010) and a mid-Miocene surface mean AHE ages on low-eU basement samples in Cooling History of Samples in the Western annual temperature of 13–18 °C (discussed later the gorge are generally younger than the AFT Grand Canyon Region herein), we can calculate a paleogeothermal gra- ages, consistent with residence near 60 °C for dient of 25 ± 3 °C/km, using the stated limits on much of Tertiary time, as suggested by the AFT Apatite Fission-Track Ages depth and surface temperature as a conservative modeling. The gorge samples yield a strong cor- AFT ages from western Grand Canyon estimate of error. relation between eU and age. Using a radiation- (Naeser et al., 1989; Kelley et al., 2001) and A somewhat lower value of 20 °C/km was re- damage model of the age-eU dependence, the the Azure Ridge–Gold Butte area in the South ported in two recent studies, primarily because samples resided near 55 °C between 65 and Virgin Mountains (Fitzgerald et al., 1991, 2009) they both assumed that the thickness of the 20 Ma, at which time they cooled rapidly from provide constraints on the cooling history not Phanerozoic section above the basal Cambrian >50 °C to <32 °C (Flowers et al., 2008). The only of basement rocks at the bottom of Grand unconformity lies in the range 3500–4000 m, AHE data thus independently confi rm the two- Canyon, where the time of incision is at issue, rather than the 2500 m value used here (Bernet, stage cooling history suggested by the AFT data, but also on nearby basement rocks overlain by at 2009, p. 182; Fitzgerald et al., 2009, p. 15). A and refi ne the timing of cooling to near-surface least 2500 m of Paleozoic–Tertiary overburden value of 3500–4000 m is the correct value for temperatures to be ca. 20 Ma (point B, Fig. 4A). until 17 Ma (e.g., Brady et al., 2000). the total Cambrian through Jurassic section, in- In marked contrast to the AFT data, AHE In the vicinity of the mouth of Diamond cluding a relatively thin section of disconform- data from samples on the canyon rim near the Creek, the river bends 135° from a long ably overlying mid-Tertiary strata (Brady et al., crest of the Kaibab arch north of Grand Can- S-fl owing reach that transects the Coconino ter- 2000, p. 1379). However, in the fault blocks yon yield the same ages and age-eU curves as race to a NW-fl owing reach parallel to the strike comprising the area from which the thermo- the basement samples in the gorge, which are of the Arizona homocline (Figs. 1 and 2). There, chronometric data were collected (Azure Ridge 1500 m lower in elevation (point A, Fig. 4A). the AFT age in basement rocks near the basal and environs, Fig. 2), the Tertiary unconformity These data also suggest residence near 55 °C Cambrian unconformity is 75 ± 5 Ma, with is located near the basal Mesozoic unconformity, from 65 to 20 Ma, followed by rapid cooling a mean track length of 14.0 µm and relatively omitting some 1200 m of Triassic and Jurassic to near-surface temperatures at 20 Ma (Flowers tight clustering (Fig. 4B, point D; Kelley et al., strata present farther north in the Virgin Moun- et al., 2007, 2008). South of the canyon, apa- 2001). The best-fi t thermal models of age and tains. Estimates of the distances between vari- tites from four samples on the plateau did not track-length distribution of this sample suggest ous PAZ and PRZ boundaries in the basement yield systematic age-eU variations for thermal rapid cooling from >110 °C to <65 °C in Cam- rocks, which do not depend on the estimate modeling, but as for samples within and north of panian time, followed by residence somewhere of sedimentary thickness, strongly support the canyon, the youngest ages in these samples between 20 and 50 °C after 60 Ma (Kelley et al., the 20 °C/km fi gure (Bernet, 2009; Fitzgerald are consistently quite young and range from 2001). AFT ages of samples 25 and 50 km up- et al., 2009). The disparity likely refl ects a sub- 28 to 18 Ma, indicating that they had reached stream to the north are younger (61 and 46 Ma, tle but real increase in the gradient within the near-surface temperatures during that interval respectively; Kelley et al., 2001). Although no upper 5 km of the crust relative to that below, (Fig. 4A; Flowers et al., 2008). AHE age dating or thermal modeling has been as predicted by model geotherms for the south- Even though the AHE data indicate that the done on these samples, they appear to have ther- western United States (e.g., Lachenbruch and sub–65 °C cooling histories of gorge and rim mal histories more similar to the eastern Grand Sass, 1978, their fi g. 9–5). Hence, for the pur- samples are similar, in the Kaibab arch area, the Canyon than to the sample near Diamond Creek. poses of estimating erosion in the uppermost difference in low-eU AHE and AFT ages is only In the Azure Ridge area of the South Virgin 3–5 km of the crust in this region, a gradient 45 Ma at the bottom of the canyon and >100 Ma Mountains, ~10–20 km NW of the western ter- of 25 °C/km is appropriate (e.g., Quigley et al., on the rim. This difference refl ects pre–80 Ma minus of Grand Canyon (Fig. 2), basement AFT 2010), whereas for estimates of the timing and residence of the basement samples in the gorge ages from the steeply tilted fault block are consis- amount of tectonic denudation for large fault near the base of the fi ssion-track PAZ, versus tently 17–15 Ma, up to a position 1–2 km struc- blocks, values near 20 °C/km are more appro- that of rim samples, which were at least 1500 m turally below the basal Cambrian uncon formity, priate (Fitzgerald et al., 2009). above the base of the PAZ. Thus, rim and gorge with tightly clustered track-length distributions About 40–50 km north of the western termi- samples (points A and B, Figs. 4A and 4C) that >14.0 µm (Fitzgerald et al., 1991, 2009). Ages nus of the Grand Canyon in the North Virgin were at much different temperatures prior to are progressively older structurally upward Mountains, extensive exposures of basement 80 Ma, refl ecting their different paleodepths, from this position, reaching 50 Ma just below rocks along and beneath the basal Cambrian converged in temperature and accordingly the unconformity, with diffuse track-length dis- unconformity yield ages ranging from 23 to had a common depth and thermal history after tributions ranging from 12.0 to 13.2 µm (point 10 Ma, i.e., much younger than ages in base- 70 Ma. These data suggest that Campanian– C, Fig. 4B). In the area best constrained struc- ment rocks in this position in the South Virgin Maastrichtian erosion of 1500 m in the canyon turally (northernmost Azure Ridge area, Fig. 2; Mountains (Quigley et al., 2010). Most of the indicated by the AFT data was primarily inci- Brady et al., 2000), sample locations indicate ages range from 17 to 14 Ma. For these samples, sion downward through Mesozoic strata, creat- that the 110 °C isotherm resided somewhere nine measured mean track lengths range from ing a canyon of roughly the same depth as the between 1000 and 1600 m below the basal 13.7 to 14.7 µm, averaging 14.0 µm (Fig. 4B). modern one, cut in younger strata now eroded Cambrian unconformity until 17 Ma, when the For an older group of ages from 23 to 19 Ma, away (Flowers et al., 2008). This event was fol- fault block was rapidly upended by extensional mean track lengths on fi ve samples range from lowed by a second pulse of unroofi ng in late faulting. The 2500-m-thick Paleozoic–Tertiary 12.7 to 13.8 µm, averaging 13.2 µm. Based on

8 Geological Society of America Bulletin, Month/Month 2010 Geological Society of America Bulletin, published online on 26 January 2011 as doi:10.1130/B30274.1

California River structural position and track-length modeling, the Hualapai Plateau (Fig. 2; e.g., Elston and The fact that the basement AFT and AHE the base of the PAZ in these samples lay <400 m Young, 1991). These two samples yielded an ages are about the same in the Diamond below the basal Cambrian unconformity at uncertain age of 89 ± 7 (Separation Point batho- Creek area places strong constraints on the 17 Ma, such that deeper samples yielding lith) and a better constrained age of 71 ± 3 Ma post–70 Ma (post-Campanian) thermal his- the younger group of ages were completely (245-Mile pluton, Fig. 5; Flowers et al., 2008). tory. Because these apatites were above 110 °C annealed prior to rapid mid-Miocene unroofi ng, Based on the proximity of these three samples, at 75 Ma and cooled quickly below 65 °C on and samples within a few hundred meters of and the similarity of their positions relative the basis of the AFT track-length data, they the unconformity were acquiring tracks during to the Colorado River, Coconino terrace, and were completely annealed prior to cooling, relatively slow erosion in Oligocene and early Paleo gene unconformity, the best estimate of and hence any damage would had to have oc- Miocene time (Quigley et al., 2010). These data the time of cooling below 70 °C in this area is curred after 70 Ma. Thus, although some dam- confi rm the presence of additional overburden 71 ± 3 Ma, or late Campanian to early Maas- age has occurred, these samples have low eU in the North Virgin Mountains relative to the trichtian time. This age is 4 m.y. younger than, values and exhibit no dependence of age on eU South Virgin Mountains prior to Miocene un- but overlaps within one standard deviation, the (Fig. 5), as would be expected if post–70 Ma roofi ng, as expected from the presence of strata AFT age (point D, Fig. 4B), and overlaps within damage was having a signifi cant effect on age as young as Jurassic below the basal Tertiary error the estimated age of rapid cooling in the (Flowers et al., 2007, 2008). unconformity in the area. They also confi rm an Kaibab arch area of 75 ± 6 Ma (Dumitru et al., For apatites without signifi cant radiation estimate of the geothermal gradient at 17 Ma 1994; Kelley et al., 2001). damage, residence at temperatures within the in the upper 5 km of the crust of 25 °C/km In the South Virgin Mountains, all AHE PRZ results in signifi cant He loss and propor- (Quigley et al., 2010; Fig. 4B). ages are 15–17 Ma, the shallowest of which is tionate reduction in AHE age. The difference 900 m below the basal Cambrian unconformity. between the AFT age and the best constrained Apatite (U-Th)/He Ages Because the base of the mid-Miocene PRZ for AHE age is (75 ± 5) – (71 ± 3) = 4 ± 6 m.y. The AHE age from a sample at the mouth apatite is ~1300 m below the unconformity The difference between the AFT age and the of Diamond Creek (Diamond Creek pluton) and the geotherm is 25 °C/km, the base of the mean age of the three AHE ages in the area is 75 ± 10 Ma (Fig. 5; Flowers et al., 2008). mid-Miocene PRZ was located near the uncon- (treating the 13 single-grain AHE ages as a Although uncertain, it is concordant with the formity (point C, Fig. 4B; Reiners et al., 2000; single population) is (75 ± 5) – (75 ± 4) = 0 75 ± 5 Ma AFT age determined by Kelley et al. Fitzgerald et al., 2009). ± 6 m.y. Clearly, more accurate data would (2001) from the same location, 25 m.y. older be desirable. Nonetheless, the distributions of than the AFT age near the unconformity in the Comparison of the Diamond Creek Area with single-grain age determinations from three dif- South Virgin Mountains and 55 m.y. older than the South Virgin Mountains ferent igneous bodies in the area with low but AFT ages near the unconformity in the North Just as the estimate of pre-incision tem- variable eU (Fig. 5) are suffi cient to show that Virgin Mountains. Regardless of the precise age perature difference between samples at dif- further analyses would be unlikely to produce of this sample, it is consistent with rapid cool- ferent structural levels provides an estimate of a population of AHE ages that, conservatively, ing of the area below 65 °C in the Late Creta- the depth of incision in eastern Grand Canyon average less than 60 Ma. ceous, as indicated by track-length modeling (Fig. 4C, curves A and B prior to 70 Ma), the Helium loss in apatite with the low eU (Kelley et al., 2001). Ages from two additional postincision temperature difference between (<14 ppm) and relatively small grain radii samples at river level ~15–20 km downstream two samples on a common datum also provides (<60 µm) of the 245-Mile pluton is rela- from the mouth of Diamond Creek were also a basis for estimating the depth of the canyon tively severe, even at the cooler end of the obtained (Fig. 2; Flowers et al., 2008). These (Fig. 4C, curves C and D after 70 Ma). Collec- PRZ. In forward models using the radiation- samples occupy a structural position similar to tively, the data indicate that for the westernmost damage and annealing model (RDAAM) of the Diamond Creek sample, lying along the SW Grand Canyon region, the temperature at the Flowers et al. (2009) and the HeFTy software boundary of the Coconino terrace. As discussed unconformity beneath the plateaus adjacent to (Ketcham et al., 1999), apatites are cooled further later herein, both of these samples and the Grand Canyon at 17 Ma, and presumably for over a 10 m.y. period from 120 °C down to a the Diamond Creek sample lie less than 10 km much of Tertiary time, was ~70 °C. Apatites at temperature ranging from 20 °C to 50 °C, after north of exposures of the unconformity between the mouth of Diamond Creek were clearly well which they are held at a constant temperature Paleozoic strata and Paleogene Rim gravels on below 70 °C at this time, but how much lower? for a period of 70 m.y. The models predict that apatites retain nearly all He if they reside below 30 °C, but lose a substantial fraction of it above 30 °C (Fig. 6). The model ages de- 50 pend strongly on the precise input parameters Diamond Creek pluton for the RDAAM model. For apatites residing 40 Figure 5. Plot of single-grain 75 ± 10 Ma between 30 and 40 °C, predicted ages vary by apatite (U-Th)/He ages versus 30 about ±10% over a range of acceptable input effective uranium concentration eU parameters based on experimental calibration for three Lower Granite Gorge Separation Point (ppm) 20 batholith (Fig. 6; see table 1 and fi g. E4–4A in Flowers basement samples (data from 89 ± 7 Ma et al., 2009). For the 245-Mile pluton apatites, Flowers et al., 2008). Sample 245-Mile pluton 10 71 ± 3 Ma the oldest ages predicted by the model drop locations are on Figure 2, keyed below 60 Ma at a residence temperature of by age. 0 35 °C (Fig. 6), which thus represents a con- 04080 120 servative upper bound on the post–70 Ma resi- (U-Th)/He age (Ma) dence temperature of the sample.

Geological Society of America Bulletin, Month/Month 2010 9 Geological Society of America Bulletin, published online on 26 January 2011 as doi:10.1130/B30274.1

Wernicke

Figure 6. Curves showing apa- Bidahochi AHE ages of 18–26 Ma along the SE tite (U-Th)/He model age as 80 margin of the area of Bidahochi deposition, con- a function of temperature for Minimum age of low eU apatites, fi rming that kilometer-scale erosion affected at Lower Granite Gorge ca. 60 Ma samples cooled rapidly from least this part of the basin between Chuska and

120 °C and held isothermally for 60 Bidahochi time (Flowers et al., 2008). 70 m.y., for apatites with prop- Additional constraints bearing on mid- to erties similar to the 245-Mile late Tertiary elevation of the plateau are based pluton sample (Figs. 2 and 5; on estimation of depositional temperatures of eU = 14 ppm; grain radius = 40 lake sediments in the Bidahochi Formation, and 60 µm), using the radiation 2 comparison with temperatures of paleolakes 4 damage and annealing model 3 1 in adjacent lowlands (Huntington et al., 2010). (RDAAM) of Flowers et al. (U-Th/He) apatite age (Ma) age apatite (U-Th/He) Temperatures were determined using carbonate 20 (2009). Curves labeled 1–4 cor- “clumped-isotope” paleothermometry, a new respond to parameter sets 1–4 Tmax ~35 °C tool based on the relative enrichment of bonds given in table 1 of Flowers et al. between heavy isotopes (13C–18O) in calcite and (2009). All parameter sets pre- 0 dolomite. This enrichment is a function of crys- dict ages younger than 60 Ma 20 30 40 50 tallization temperature, where development or for residence temperatures Residence temperature (°C) “clumping” of heavy-isotope bonds decreases above 35 °C. with temperature. A key advantage of this ther- mometer is that it is not dependent on assump- tions about the isotopic composition of oxygen or carbon in surface waters (Eiler, 2007). To

This result must be qualifi ed by the fact that CENOZOIC ELEVATION AND measure enrichment, CO2 gas is evolved by acid the radiation-damage models are least certain CLIMATE ON THE SOUTHERN digestion of the sample and then purifi ed. The at very low temperatures, which are farthest COLORADO PLATEAU measured quantity is the relative abundance of outside the range of experimental conditions CO2 isotopologues (molecules of the same com- on which the models are based (K.A. Farley, Stratigraphic and thermochronologic evi- position but differing mass) which contain one 2009, oral commun.). Therefore, although avail- dence for ~1400 m of Late Cretaceous relief heavy isotope of oxygen and one of carbon. This able data and models confi rm that temperatures in the western Grand Canyon region and some isotopologue has mass 16 + 18 + 13 = 47, and were below 35 °C for most of Tertiary time (e.g., 1500 m in eastern Grand Canyon suggest mini- therefore the measured quantity is denoted as Δ Flowers et al., 2008), further studies of He dis- mum Late Cretaceous elevations of at least 47 (Ghosh et al., 2006; Eiler, 2007). Measure- tribution in Lower Granite Gorge apatites (e.g., these amounts for portions of the southwestern ments from both modern and ancient samples using the 4He/3He method; Shuster and Farley , plateau (Young, 1979; Flowers et al., 2008). deposited near sea level in the lower Colorado 2004) will be necessary to better quantify limits The apparent stability in the thermal histories of River basin were compared with the Bidahochi on the late Tertiary thermal history, which may both regions through most of Tertiary time hints Formation, providing an opportunity not only to include scenarios where the samples were that elevation of this magnitude was maintained estimate changes in elevation of high-elevation somewhat warmer than 35 °C. throughout the period. samples relative to sea level, but also to quan- The salient point made by the concordance After widespread Paleocene–Eocene aggra- tify the infl uence of climate on depositional of the AFT and AHE ages is that if the only dation of fl uvial and lacustrine deposits on the temperatures. Δ information available were the AHE ages, then plateau, but before integration of the upper and Measurements of 47 of modern lacustrine cal- they could be interpreted in terms of prolonged lower Colorado (the period from ca. 36 to 6 Ma), cite deposited from 350 to 3300 m elevation in residence in the PRZ beginning at a time much the plateau most likely experienced modest ag- the southwestern United States suggest a lacus- earlier than 70 Ma, at temperatures below 70 °C gradational and erosional events, with the most trine carbonate temperature (LCT) lapse rate but substantially higher than 35 °C, as is the pronounced erosion along the SW fl ank. This of 4.2 °C/km with a zero-elevation intercept of case for the eastern Grand Canyon region. The history is best illuminated in the central and 24 °C (Fig. 7). Limestones from the Bidahochi fact that the temperature dropped soon before southern plateau. There, aggradation of as much Formation (~1900 m above sea level) record the recorded AHE age precludes this hypoth- as 500 m of eolianites of the Chuska erg, down- temperatures of 22–25 °C, 8 °C warmer than esis. If correct, it in turn suggests a temperature wind from Oligocene volcanic constructions the temperature predicted by the modern LCT difference of at least 35 °C existed between along the southeastern margin of the plateau, trend, perhaps implying signifi cant post–6 Ma “plateau” (i.e., South Virgin Mountains) and occurred from 34 to 25 Ma, followed by their uplift. However, limestones measured in the Lower Granite Gorge basement samples at removal and further incision of a few hundred low-standing Colorado River trough (Bouse 17 Ma (Fig. 4C, curves C and D). The tempera- meters into their substrate (Cather et al., 2008). Formation and Hualapai Limestone) cluster ture difference corresponds to a canyon depth The erosion surface at 16 Ma beneath the Bida- around an average value of 30.5 °C, except of no less than (35 °C)/(25 °C/km) = 1400 m. hochi Formation (Fig. 1) is as much as 1200 m for two samples from the southernmost part of This estimate is nearly all of the modern canyon lower in elevation than the highest preserved the basin (Fig. 7). These samples yielded tem- depth of 1500 m, and similar to the minimum lo- erg deposits ~100 km to the east near the center peratures 6–8 °C cooler, which are interpreted cal paleorelief preserved beneath the Rim gravel of the plateau, suggesting that erosion between to result from the infl uence of a nearby marine unconformity nearby on the Hualapai Plateau 25 and 16 Ma, including that of the erg depos- climate on lake surface temperatures (Hunting- (Elston and Young, 1991), discussed further in its, could have been more than 1000 m (Cather ton et al., 2010). Linear regression of all the an- the following sections. et al., 2008). This fi gure is supported by sub- cient samples except for the southernmost group

10 Geological Society of America Bulletin, Month/Month 2010 Geological Society of America Bulletin, published online on 26 January 2011 as doi:10.1130/B30274.1

California River

Bouse and Hualapai samples the 17 Ma surface MAT on the westernmost pla- 88–646 m teau. Assuming the surface was at 2000 m eleva- 35 Late Miocene LCT (n = 11): tion, surface MAT in the Miocene would have T (°C) = –4.1 °C/km × elev + 32.1 °C been the present value of 10 °C plus 3–8 °C to 30 adjust for cooling of the climate. Bidahochi samples High elevation of the plateau during Muddy 1806–1898 m Creek/Bidahochi time is further supported 25 climate-adjusted modern LCT by the depositional geometry of the Muddy Creek Formation against the Grand Wash Cliffs 20 (Fig. 3). Near the western terminus of Grand ΔT = –7.7 ± 2.0 °C Canyon, Muddy Creek strata lie in buttress un- 15 conformity against the cliffs. The Muddy Creek Temperature (°C) modern LCT was probably not deposited substantially below 10 sea level (cf. Lucchitta, 1979), as suggested by (1) the similarity in depositional temperatures between the Hualapai Limestone and Bouse 5 0 1000 2000 3000 Formation (Fig. 7; Huntington et al., 2010); Elevation (m) (2) the requirement of north-to-south hydro- logical fl ow between the Hualapai and Bouse Figure 7. Carbonate clumped-isotope thermometry temperature basins (e.g., Spencer et al., 2008a); and (3) the estimates versus modern elevation for samples collected in the rarity of active nonmarine deposition below sea Colorado River basin (after Huntington et al., 2010). Data points level. The elevation difference between the low- marked by unfi lled circles are interpreted to refl ect cooling of lake est exposures of Muddy Creek at river level and surface temperatures due to the infl uence of marine climate near the Shivwits Plateau immediately east of Grand the late Miocene delta. Inland samples defi ne a lacustrine carbonate Wash Cliffs is 1600 m (Fig. 3), which repre- temperature (LCT) lapse rate for the late Miocene similar to the sents a minimum estimate of the elevation of modern LCT lapse rate, suggesting little or no elevation adjustment the west-central margin of the plateau in mid- of the southwestern Colorado Plateau since 16 Ma. Vertical arrow Miocene time. shows apparent 8 °C of cooling of lake depositional temperatures since 6 Ma, interpreted to refl ect climatic cooling. PROVENANCE OF LATE CRETACEOUS–PALEOGENE DEPOCENTERS yields a zero-elevation intercept of 32 °C and a 1000 m of uplift (Sahagian et al., 2003; see dis- paleo-LCT lapse rate of 4.4 °C/km (Fig. 7). cussion in Huntington et al., 2010), but they are Grand Canyon Region Because the mean annual temperature (MAT) consistent with the hypothesis that kilometer- lapse rate is relatively insensitive to climate scale relief in the southwestern portion of the Rim Gravels South of Grand Canyon change under generally arid conditions at plateau had developed by latest Cretaceous time The Rim gravel deposits are typically ~100 m middle latitudes, these data suggest a history (Flowers et al., 2008; Huntington et al., 2010). thick on the plateau, and are preserved to within of little or no elevation change for any of the The 8 °C cooling in the LCT curve may over- 50 km SW of the rim of eastern Grand Canyon samples since 6 Ma. The 8 °C difference in estimate the change in MAT since Miocene and to within <10 km of the Colorado River in temperature between the modern and ancient time, if seasonality were more pronounced then. western Grand Canyon (Figs. 1 and 2; Elston samples at any given elevation suggests that the Lacustrine carbonate temperatures primarily and Young, 1991; Young, 2001a). They contain Miocene climate, at least as revealed in lake- refl ect late spring/early summer lake-surface clasts of volcanic rocks with predominantly surface temperatures, was substantially warmer temperatures, so it is possible that a signifi cant Late Cretaceous K-Ar ages (mainly 80–64 Ma; than today’s climate, an observation consistent component in the rise of LCT temperatures Elston et al., 1989). They also contain detrital with the glacial conditions of the Quaternary could occur without commensurate increase in grains with AHE ages as young as 50 Ma, and versus the nonglacial conditions of the Mio- MAT (Huntington et al., 2010). Further, ongo- lie on a substrate with AHE ages as young as cene, as well as other proxies for paleoclimate ing studies of low-elevation modern lakes in 53 Ma. Thus, on the basis of thermochrono- in the western interior (e.g., Zachos et al., 2001; the region suggest a somewhat higher modern logical data alone, deposition occurred in early Cather et al., 2008; Chapin, 2008; Young, 2008, LCT lapse rate and sea-level temperatures than Eocene or later time (Flowers et al., 2008). and references therein). Further, the consistency estimated from the more limited data set of Lacustrine carbonates within the Rim gravels of temperatures between 16 Ma limestones and Huntington et al. (2010; J. Thompson, J. Eiler, contain a nonmarine invertebrate fauna simi- 6 Ma limestones within the Bidahochi Forma- 2010, personal commun.). Ocean-surface tem- lar to that of the well-dated Paleogene fl uvial tion suggests at best only small changes (<2 °C) peratures through much of Tertiary time were and lacustrine strata along the NE fl ank of the since 16 Ma; this is also consistent with the con- ~3 °C warmer than during the Quaternary (e.g., Arizona homocline in southern Utah and else- trast between generally nonglacial times and the Zachos et al., 2001). Given the continental set- where in western North America (Young, 1999). Quaternary. These results are not consistent with ting of the plateau, conservative limits on the On this basis, they have been regarded as early surface uplift estimates based on basalt vesicle Tertiary MAT would, therefore, be between 3 Eocene in the Long Point area (Fig. 1) and pos- altimetry on Pliocene basalts on the plateau in and 8 °C warmer than today. These data provide sibly as old as Late Cretaceous in other areas east-central Arizona, which suggest more than a basis for the estimate in the previous section of (Young, 1999, 2008). Thus, in combination

Geological Society of America Bulletin, Month/Month 2010 11 Geological Society of America Bulletin, published online on 26 January 2011 as doi:10.1130/B30274.1

Wernicke with the apatite He data, the depositional age is Hindu, and Peach Springs canyons (Fig. 2), are in Peach Springs Canyon lie in a fault block early Eocene (Flowers et al., 2008). Young and locally cut very close to the sub–Rim gravel along the trace of the Hurricane fault and are McKee (1978) suggested that the basal uncon- unconformity (Elston and Young, 1991; Young substantially lower than exposures east and formity on the Rim gravels represents a Late 2001a; Fig. 8). west of the fault zone, which lie at ~1300 m Cretaceous–Paleogene drainage network up- The lowest preserved exposures of the Peach elevation (Billingsley et al., 1999), and hence stream from the Paleogene deposits in southern Springs–Truxton paleocanyon come to within movements along the fault zone during both Utah. Based on radiometric ages in Rim grav- 8 km of Grand Canyon, where both the modern late Cenozoic and Laramide time complicate els in the Fort Apache region (Potochnik and and ancient channels drain to the NNE (Figs. 2 estimates of the position of the river channels Faulds, 1998) and on paleontological grounds, and 8; Billingsley et al., 1999). The bottoms of relative to the modern Colorado River grade a however, Cather et al. (2008) regarded the Rim the ancient channels lie at modern elevations of short distance to the north. Given the modern gravels in the Hualapai Plateau area as part of between 830 and 1000 m, and now slope very elevation of the lowest gravels and the eleva- a plateau-wide period of aggradation of late gently SW, opposite to the NE depositional tion of gravels east and west of the fault zone Eocene to early Oligocene age, as described in paleo slope (Young, 2001a). A minimum of as bounds on the position of the ancient chan- the previous section. 0.8° of SW tilting (opposite the NE dip of the nels, they would intersect the Colorado at a Adjacent to the Arizona Transition Zone in Hualapai Plateau Paleozoic section) after depo- modern elevation between 1050 and 1350 m at the Peach Springs area, the gravels fi ll paleocan- sition of the gravel is necessary for the base of the mouth of Diamond Creek (Figs. 2 and 8). yons incised deeply into the Hualapai Plateau the channel network to have been initially hori- The projected intersection is as much as 400 m (Figs. 2 and 8; Young, 1979). Prior to Rim gravel zontal, taking into account some 170 m of late below the mean elevation of the Hualapai Pla- deposition, the paleocanyons had incised down- Cenozoic offset on the Hurricane fault zone in teau (vertical measurement 2, Fig. 8) and as ward to create local relief of at least 1200 m the area (Fig. 8; Young, 2001a). much as 750 m below the Shivwits Plateau im- (vertical measurement 1, Fig. 8). Net aggrada- The existence of a Paleogene stream chan- mediately to the north of Grand Canyon (verti- tion of as much as 300 m occurred between the nel within a few kilometers of the Colorado cal measurement 3, Fig. 8; Young, 1985). This early Eocene and the eruption of 19 Ma basalts River, which in modern exposures is cut to constraint appears to limit any post–36 Ma in- near the top of the section (Separation Canyon within 600 m of the modern river grade, speaks cision of Grand Canyon to between 700 and basalt, Fig. 8). Paleocurrent indicators and the to the antiquity of portions of the modern 1000 m at Diamond Creek (vertical measure- provenance of the gravels indicate overall NE landscape (Young, 1979, 2008; Elston and ment 4, Fig. 8). However, as described already, transport, from the Arizona Transition Zone to- Young, 1991; Karlstrom et al., 2007), and is a residence temperature of <35 °C for basement ward the plateau (Young, 1966). Since 19 Ma, problematic for any model wherein western samples suggests that even with a minimum of these channels were variably reincised, reexpos- Grand Canyon incision is primarily younger 700 m of post-Eocene erosion, substantial He ing their bottoms. Modern southern tributaries than 6 Ma (e.g., Lucchitta and Jeanne, 2001). loss would have occurred after 70 Ma, which to western Grand Canyon, including Milkweed, The Rim gravel exposures at 1000 m elevation is treated in more detail in following sections.

Grand BEND SSW Hualapai Plateau Shivwits Plateau NNE Music Mountains Canyon Tb 2 Separation Canyon 750 m (3) Pk basalt (19 Ma) 1200 m (1) Hindu Canyon paleochannel 400 m (2) 1 1300 m (5) MP 1 2 700 - 1000 m (4) Truxton Valley wells PEACH SPRINGS-TRUXTON CD

Elevation (km)Elevation (projected) PALEOCANYON Grand Canyon base level after 70 Ma Xg

0 20 40 60 Horizontal distance (km) VERTICAL EXAGGERATION = 5:1 Neogene valley fill Paleoflow direction, Rim gravels 10° 5° Paleogene Rim gravels 750 m (3) Vertical measurements discussed in text

Hypothesized Rim gravel outlet to: 1 southern Utah, via Grand Canyon upstream of Diamond Creek (e.g., Young and McKee, 1978) 0° Dip 2 southern California, via Grand Canyon downstream of Diamond Creek (this report)

Figure 8. Cross section through the western Grand Canyon region showing geometric relationships between various geomorphic and geologic features discussed in text (based on Young, 1982). Xg—Proterozoic crystalline basement; CD—Cambrian and Devonian strata; MP—Mississippian to Permian strata; Pk—Permian Kaibab Formation; Tb—8 Ma Shivwits basalts. Vertical arrows show measurements of elevation differences discussed in text. Hindu Canyon channel depicted assuming present minimum elevation of Rim gravel deposits in Peach Springs Wash (1000 m) has not been modifi ed by faulting (see text for discussion).

12 Geological Society of America Bulletin, Month/Month 2010 Geological Society of America Bulletin, published online on 26 January 2011 as doi:10.1130/B30274.1

California River

Comparison of Rim Gravels with “Canaan transport of far-traveled sands and gravels was zircon age peak from a sample from the lower Peak–Type” Gravels North of Grand Canyon replaced by accumulation of locally derived siliciclastic part of the Claron is Late Jurassic Numerous undated gravel deposits exposed material, recycled from underlying formations, (>150 Ma), suggesting the return of a relatively north and west of Grand Canyon provide a in basins trapped within gentle Laramide syn- distant source, most likely Cordilleran/Rocky potential link between the Rim gravels and clines (Goldstrand, 1994; Larsen, 2007). This Mountain sources to the north and west rather coeval deposits in southern Utah (e.g., Elston was followed by time-transgressive deposition than California sources to the SW (Larsen, et al., 1989). Reconnaissance studies of these of post-tectonic basal sands and gravels of the 2007; Link et al., 2007; Larsen et al., 2010). “Canaan Peak–type” gravels indicate impor- Claron Formation from late Paleocene to middle Along the northern margin of the Colorado tant contrasts with the Rim gravels (e.g., Hill Eocene time, which exhibit evidence of overall Plateau in the Uinta Basin, detrital zircons in the and Ranney, 2008). Rim gravels are typically southward transport (Fig. 9B). The sands and Paleocene–Lower Eocene Colton Formation ex- cemented and preserved intact, and have a vari- gravels are succeeded by deposition of lacus- hibit a signature similar to the Kaiparowits For- ably rounded, diverse clast assemblage. It in- trine marl (red Claron) and then limestone mation, with derivation almost exclusively from cludes Proterozoic basement, Paleozoic cover (white Claron) that prevailed through at least the arc (Davis et al., 2010), which is in turn quite rocks, and Late Cretaceous volcanic rocks that middle Eocene time and possibly into the late similar to the signature from a sample of the record unroofi ng of the Arizona Transition Zone Eocene (Goldstrand, 1990, 1994). Cretaceous portion (post–79 Ma) of the McCoy and areas to the SW, via a NE-fl owing drainage The pre–Canaan Peak Cretaceous strata Mountains Formation in west-central Arizona network (e.g., Young, 1999, 2001b). Canaan contain detrital zircon populations for which (Davis et al., 2010; Spencer et al., 2011), located Peak–type gravels are typically exposed as the youngest ages are about the same as the within the Mojave/Mogollon Highlands region. unconsolidated lag deposits on the modern depositional age through most of the Cam- Thus, in marked contrast to the Maastrichtian erosion surface, and consist primarily of vari- panian epoch. Three stratigraphic levels yield, cutoff in the supply of arc-derived material in colored, well-rounded quartzite pebbles and from bottom to top, minimum age peaks at 82, the southern Utah basins (Larsen et al., 2010), cobbles, interpreted by Hill and Ranney (2008) 77, and 73 Ma, indicating strong input from the northern margin of the plateau continued to to be recycled from the Maastrichtian–early the active Cordilleran magmatic arc to the SW receive arc detritus. Paleocene Canaan Peak Formation of southern (Larsen, 2007; Jinnah et al., 2009; Larsen et al., Utah on the basis of their macroscopic appear- 2010). The youngest of these levels, the Kai- Strata Downstream of Grand Canyon in ance. The quartzite clasts in the Canaan Peak parowits Formation, has an unusual signature. Southern California contain distinctive chert litharenite and black In contrast to enveloping strata, which contain chert clasts containing microfossils that can be large components of Paleozoic and Precam- Tectonic Setting and Depositional Substrate directly linked to the Mississippian Eleana For- brian zircon, the Kaiparowits is predominantly Tectonic reconstruction of the right-lateral mation to the west in southern Nevada (Gold- arc-derived. San Andreas transform restores the distinctive strand, 1990). Because it is possible that these Surprisingly, following this pulse of sediment latest Cretaceous–Paleogene depocenters of the deposits are entirely post–6 Ma in age, they derived from the active arc, the youngest zircons Western Transverse Ranges of southern Califor- do not as yet place any fi rm constraints on the in the overlying Canaan Peak are 103 Ma, in- nia and Salinian terrane in central California to paleohydrology of the region prior to develop- dicating that the active-arc source was cut off a position due SW of the Grand Canyon region ment of the modern Colorado River. from the region by Maastrichtian time (Larsen, at 6 Ma and earlier times (e.g., Howard, 1996; 2007; Link et al., 2007; Larsen et al., 2010). The Atwater and Stock, 1998). The restored belt of Strata Upstream of Grand Canyon in Utah 103 Ma age, combined with the onset of deposi- these terranes is 500 km long, approximately tion of coarser material, strongly suggests deri- aligning it with the 500-km-long Arizona Tran- In southern Utah (Kaiparowits Plateau and vation of at least part of the Canaan Peak from sition Zone and environs (Figs. 1 and 9; e.g., environs, Fig. 1), extensive exposures of Upper sources less distant than the active arc, including Saleeby, 2003; Jacobson et al., 2010). Cretaceous to Eocene deposits provide a detailed the mid-Cretaceous Delfonte volcanics in the After reconstruction, to the NW and SE of record of the evolution of drainage and tec- eastern Mojave region of southern California the Transition Zone, a broad Cretaceous forearc tonism. Through Campanian time, these deposits (Goldstrand, 1990; Larsen, 2007). basin is well preserved to the west of the Cor- record primarily fluvial deposition of far- A sparse component of Campanian zircon dilleran arc, including a concordant Upper Ju- traveled sand and mud, with alternating episodes is found in the Paleocene–lower Eocene Pine rassic through Eocene section as much as 15 km of NE (longitudinal to the Sevier orogen) and SE Hollow Formation. These few grains (only fi ve thick (e.g., Ingersoll, 1982; Fig. 9B). In the (transverse) transport (Fig. 9A; Lawton et al., total), nonetheless, exhibit an age progression Western Transverse Ranges and Salinian ter- 2003). These deposits have variable sources, in a reversed order from that observed in the rane, there is a stark contrast. Pre-Maastrichtian including detritus from the Sevier orogen and a Campanian section, becoming progressively deposits are either thin or absent, and a concor- high plateau region immediately to its west in older from bottom to top, with ages of 71, 78, dant Maastrichtian through Eocene section as Nevada and Utah (“Nevadaplano” of DeCelles, and 81 Ma (Larsen, 2007). Using paleocurrent much as 8 km thick rests nonconformably on 2004; Fig. 9A), and the nascent Mogollon High- data and characteristics of modern erosion pat- Late Cretaceous arc plutons and pre-Cretaceous lands in Arizona (Dickinson and Gehrels, 2008). terns in these units, the detrital zircon data from crystalline rocks as old as Early Proterozoic Their deposition was followed in Maastrichtian the Pine Hollow Formation were interpreted (e.g., Chipping, 1972; Grove, 1993). to early Paleocene time by deposition of the by Larsen (2007, his fi g. 15) to refl ect local re- The tectonic elements characterizing the Canaan Peak and Castle Gate formations, which cycling via unroofi ng from a developing anti- Cretaceous arc NW and SE of the thick Maas- contain bouldery gravels and are dominated by cline (e.g., Burbank et al., 1988). In contrast trichtian–Eocene depocenter include fi ve dis- E- to ENE-directed fl ow (Goldstrand, 1992). to the Pine Hollow Formation and all older tinct, laterally persistent belts, from SW to NE By Paleocene to early Eocene deposition of units, the post-tectonic Claron Formation did (Fig. 9A; e.g., Dickinson, 1981; Ingersoll, 1998; the overlying Pine Hollow Formation, regional not yield any Cretaceous zircons. The youngest Saleeby, 2003): (1) a subduction complex;

Geological Society of America Bulletin, Month/Month 2010 13 Geological Society of America Bulletin, published online on 26 January 2011 as doi:10.1130/B30274.1

Wernicke

Transverse Miogeoclinal C r e t a c e o u s I n t e r i o r S e a w a y drainage hinge zone

A A r c d e t r i t u s R e t r o –

“N e v a d a p l a n o” W e s t e r n F o r e a r c S u b d u c t i o n E a s t e r n GC UT CO AZ NM S e v i e r o r o g e n NV California River

Figure 9. Paleogeographic maps WG EGEG showing: (A) Campanian posi- a r c tion of California River drain- K age near the onset of incision in P a c i f i c the Grand Canyon region. Red line shows position of drain- N N age divide; black arrows show paleo current directions of the Wahweap Sandstone from Law- ton et al. (2003), along with hypo- LA t e r r a i n thetical position of drainage transverse to the Sevier orogen. a r c c o m p l e x Sevier uplands and hypothetical 80 Ma T r e n c h “Nevadaplano” high plateau are from DeCelles (2004). (B) Early b a s i n a r c Eocene positions of Arizona

River drainage, the drainage 0 500 system in the Fort Apache region km (Potochnik, 2001), and ancestral Gila and Amargosa Rivers (How- ard, 1996). Palinspastic base and Miogeoclinal hinge zone Uinta direction of plate convergence Basin are after Saleeby (2003). Position B (recycled (?) of drainage divide in Nevada arc detritus) is from Henry (2008). Abbre- Monument NV Claron viations highlighted with yellow Upwarp W e s t e r n Basin F o r e a r c background correspond to lo- S u b d u c t i o n E a s t Amargosa e r n R. Gentle GC UT CO cations in Figure 13 as follows: M o j a v e – Steep NMAZ LA—Los Angeles; N—Needles; slopes slopes K—Kingman–Lake Mead area; River EG San Juan WG—western Grand Canyon; Basin EG—eastern Grand Canyon; WG Asymmetric GC—Glen Canyon. drainage divide K P a c i f i c M o g o l l o n h i g h l a n d s

N N Ft. Apache Baca Basin region

Arizona Gila R.

Maastrichtian-mid-Miocene LAA a r c continental borderland c o m 55 Ma T r e n c h (Salinian depocenters)

b p a a

14 Geological Society of America Bulletin, Month/Month 2010 Geological Society of America Bulletin, published online on 26 January 2011 as doi:10.1130/B30274.1

California River

(2) tectonically overlying forearc basin strata; Provenance Tertiary time, and terminated in coastal basins (3) primarily Early Cretaceous tonalitic west- In the newly developed borderland basins, now preserved in the southern Coast Ranges, ern arc plutons intruding a Jurassic and older Maastrichtian strata are dominated by arc- Western Transverse Ranges, and Peninsular ensimatic arc framework; (4) Late Cretaceous derived material. Gravels at the base are angular Ranges of California. The active coastal delta granodioritic to granitic eastern arc plutons, and derived from the local basement, but they system lost connection to inland sources by the which intrude metamorphic equivalents of rocks mature upward to include well-rounded meta- formation of the San Andreas system, which behind the arc and associated volcanic strata as rhyo lite clasts and, in some sections, clasts of resulted in transfer of deposition from the now- young as 70 Ma; and (5) an ensialic retroarc metaquartzite similar to Cambrian metaquartz- offset Los Angeles area to the rapidly opening belt, which contains a depositional hinge zone ites within framework rocks of the eastern arc Gulf of California over the last 6 m.y. Howard between Paleozoic shelf deposits to the NW (Colburn and Novak, 1989; Grove, 1989). (1996) suggested that the system had two pri- (Cordilleran miogeocline) and cratonic depos- Upper Paleocene and Eocene conglomerates mary branches, one reaching to the NW from its to the SE. The hinge zone is oriented at a farther upsection record a signifi cant change the delta and tapping sources in the Death Valley high angle to the Cretaceous arc, running from in provenance. In Upper Paleocene strata ly- region (his “Amargosa-Colorado paleoriver”), just NW of Grand Wash Trough to its intersec- ing above a regional unconformity (Runyon and another reaching to the SE (“Gila-Colorado tion with the arc in the central Mojave Desert Canyon surface of Colburn and Novak, 1989) paleoriver”), tapping distinctive sources in SE (e.g., Martin and Walker, 1992; Fig. 9). Hence, and below a tuff horizon dated at 56 Ma, un- Arizona and Sonora, Mexico (“Poway-type” the modern Colorado River basin is developed metamorphosed quartz arenite clasts (ortho- clasts), as documented by Kies and Abbott mostly SE of the hinge in the cratonic portion of quartzites) become abundant, and remain (1983). Howard (1996, p. 785) raised the pos- the retroarc foreland. common in virtually all conglomerates (Col- sibility that the source of the Sespe clasts could Southwest of the Arizona Transition Zone, burn and Novak, 1989; Howard, 1996, 2000). also have been an early Tertiary paleo–Grand beginning in Campanian time, tectonic events In an upper-middle Eocene (Uintan) portion Canyon, because the petrographic match with along the continental margin severely disrupted of the Sespe Formation (Whistler and Lander, the Tapeats Sandstone is especially strong. In- the system, resulting in the juxtaposition of 2003), comparison of the petrology of ortho- deed, Grand Canyon currently contains its most the eastern arc directly against the subduction quartzite clasts with that of orthoquartzites in extensive area of exposures, and the localization complex. The juxtaposition at upper-crustal likely source regions demonstrates a strong tie of a main-stem river within a large area of out- levels is best expressed in the Coast Ranges of to the Ediacaran–Lower Cambrian Stirling, crop of Tapeats Sandstone would make its most central California along the Nacimiento fault Wood Canyon, and Zabriskie Formations to survivable subunits a prominent component of zone (e.g., Page, 1981). At deeper crustal lev- the NW of the miogeoclinal hinge zone, and to the ancient river’s bed load. However, other els, the eastern arc is underthrust by metamor- the Lower Cambrian Tapeats Sandstone on the potential sources for Tapeats clasts in eastern phosed Upper Cretaceous though Eocene trench craton (Howard, 2000). Although these units California and Arizona are possible, and follow- deposits (Pelona, Orocopia, and Rand schist occur locally within the eastern arc as meta- ing community consensus, Howard (1996) felt complexes; e.g., Saleeby, 2003; Jacobson et al., morphosed wall rocks, preservation of unmeta- the hypothesis was precluded because pre–mid- 2010). These juxtapositions have been attributed morphosed sections is largely restricted to the Miocene erosion levels in Grand Canyon had by some workers to tectonic erosion along the retroarc belt. The transition from local to ex- not reached the level of the Tapeats Sandstone Cretaceous trench, which sheared off and sub- otic, beginning in the Maastrichtian but culmi- during Sespe time. ducted the forearc and western arc, replacing it nating with the arrival of orthoquartzite clasts Restriction of the Sespe source region to with trench deposits at both shallow and deep in the Paleocene, confi rms a long-held view areas SW of the modern Colorado Plateau is crustal levels (e.g., Saleeby, 2003). An alterna- that the coastal drainages opposite the Transi- supported by comparison of detrital zircon tive model attributes the juxtaposition at shallow tion Zone expanded northeastward and became populations in late Paleozoic and early Meso- structural levels to large-magnitude, left-lateral increasingly integrated in latest Cretaceous and zoic erg deposits that are widespread on the strike-slip along the Nacimiento fault (Dickin- early Tertiary time (e.g., Woodford et al., 1968; plateau and detrital zircon populations in the son, 1983; Dickinson et al., 2005), which may Kies and Abbott, 1983; Grove, 1993; Howard, Sespe Formation. Whereas Appalachian- and also account for the southward migration of 2000). Thereafter, tectonic events associated Grenville-age zircons (ca. 0.4–1.3 Ga) are abun- deposition and metamorphism of the Pelona with the growing transform margin (Atwater dant in the erg deposits (Dickinson and Gehrels, and related schists, from >90 Ma to the north and Stock, 1998; McQuarrie and Wernicke, 2003, 2009), they are lacking in the Sespe For- to <60 Ma to the south (Jacobson et al., 2010). 2005) resulted in a cutoff of input from ex- mation (e.g., Spafford et al., 2009). Hence, if Regardless of the degree to which thrusting otic sources at ca. 15 Ma, and a return to local the Sespe headwaters did reach as far NE as the or strike-slip faulting contributed to these jux- derivation (including recycling of earlier ortho- Grand Canyon area, it would be conditional on tapositions, the Late Cretaceous nonconformity quartzites) during deposition of the Puente, the source region lacking extensive exposures of exposed at high structural levels in the eastern Monterey, and equivalent formations along the the erg deposits. arc records a dramatic transition in setting, active transform (Critelli et al., 1995; Ingersoll In this light, the modern western Grand from an earlier, inboard position along the axis and Rumelhart, 1999). Canyon would remain a potential source area, of the Cordillera, to an extending and rapidly The consistency in provenance from the east- because (1) the primary Paleozoic erg deposit subsiding continental borderland (e.g., Grove, ern arc/retroarc region from early Paleocene to (Coconino Sandstone) pinches out beneath the 1993; Saleeby, 2003; Jacobson et al., 2010). mid-Miocene time, and the fl uviodeltaic deposi- Kaibab Limestone in western Grand Canyon, The marine embayment in the continental mar- tional facies of these deposits led Howard (1996, where it is either absent or at most a few tens gin resulting from this transition (e.g., Saleeby, 2000) to conclude that a “Colorado paleoriver,” of meters thick (e.g., Wenrich et al., 1997); and 2003) began to focus drainages into the region, located at about the same place as the modern (2) Lower Mesozoic erg deposits (Navajo and resulting in thick accumulations of terrigenous Colorado River NE of the San Andreas fault related sandstones) are omitted across the un- detrital sediment (Fig. 9B). near Yuma, Arizona, existed throughout most of conformity at the base of the Rim gravels, which

Geological Society of America Bulletin, Month/Month 2010 15 Geological Society of America Bulletin, published online on 26 January 2011 as doi:10.1130/B30274.1

Wernicke

rest on pre-Navajo strata throughout the region than ~300 m, the maximum depth of inci- where Ts h70 is the surface temperature above (Fig. 1; e.g., Billingsley et al., 1999). In these sion of western Grand Canyon below the top the sample at the onset of erosion at 70 Ma, respects, western Grand Canyon is more akin of the Tapeats Sandstone. If correct, it has a and dT/dz is the geothermal gradient (a positive to the “erg-poor” Mojave/Mogollon Highlands broad range of implications for the origin of number), assumed to remain relatively steady source region than it is to the “erg-rich” source Grand Canyon and the paleohydrology of the during the small amount of slow erosion inher- areas throughout the remainder of the plateau. southwestern United States. A 70 Ma erosion ent to the problem (Fig. 10). It is noteworthy that Ediacaran–Cambrian surface, if very close to the modern one at the At present, shallow surface temperatures vary orthoquartzite clasts abundant in Sespe grav- bottom of western Grand Canyon, has the same as a function of surface elevation h according to els likely contain a substantial component of implications as would a discovery of Campanian

Grenville-age (1.0–1.3 Ga) zircons (Stewart river gravel deposits near the modern river grade. Ts (h) = Th = 0 + (dTs /dh)h, (2) et al., 2001). Thus, even though Ediacaran– The key question is, using thermochronological

Cambrian units constitute ~10%–30% of the data as a proxy, at what level in western Grand where Th = 0 is the present surface temperature at gravel fraction of the Sespe and are widely ex- Canyon can we claim such deposits once existed? sea level, and dTs/dh is the lapse rate of surface posed in the Mojave/Mogollon Highlands re- To fi rst order, given a nominal surface tem- temperature as a function of elevation (a nega- gion (Howard, 1996, 2000), they did not make perature of 20 °C and geotherm of 25 °C/km, tive number; Fig. 10). a measurable contribution to detrital zircon the upper limit of erosion is ~600 m. A fi g- The relationship between changes in surface populations recovered from Sespe sandstones ure of roughly this amount is independently air temperature and the near-surface geotherm (Spafford et al., 2009). suggested by the northward projection of the is complex (e.g., Section 2.6 in Carslaw and Peach Springs–Truxton paleocanyon, which Jaeger , 1959), but for our purposes, we note that SYNTHESIS suggests no more than 700–1000 m of erosion surface temperature changes due to long-term after 36 Ma at the latest (vertical measure- climate change ΔT (a negative number for cool- Erosion History of Western Grand Canyon ment 4, Fig. 8). ing) are small relative to the depth over which We can better estimate the post–70 Ma ero- they perturb the geotherm. This is not true for Given an apparent upper limit of 35 °C for sion history of western Grand Canyon by century-scale climate variations, which are of the temperature of western Grand Canyon base- synthe sizing (1) estimates of the mid-Tertiary order 1.0 °C and occur over a depth range of ment in the Diamond Creek area since 70 Ma, geo therm, (2) modern measurements of tem- 0–100 m, affecting temperature only modestly, it seems possible that the erosion level may perature in the shallow subsurface in the region, but affecting dT/dz by as much as a factor of 2 indeed have reached as deeply as the Tapeats (3) the difference between Quaternary and Ter- (e.g., Chisholm and Chapman, 1992). However, Sandstone, as early as Late Cretaceous time. tiary MAT, and (4) estimates of Tertiary paleo- in our case, we are dealing with a boundary tem- If so, it expands the potential reach of How- elevation in the region. perature change ΔT in the range 3–8 °C applied ard’s (2000) already extensive (~400-km-long) In the upper ~1000 m of the crust, the tem- over a time scale of at least 2 m.y. The depth Δ paleoriver systems into the SW margin of the perature Tmax at a maximum sample depth zmax of infl uence L of a boundary perturbation T plateau in NW Arizona. This hypothesis pre- through Tertiary time is scales as L ~ 2(κt)1/2, where κ is thermal con- dicts post-Campanian (post–70 Ma) erosion ductivity, and t is the time since the perturbation. –6 2 in western Grand Canyon to have been less Tmax = Ts h70 + (dT/dz)zmax, (1) For thermal conductivity of 10 m/s and t =

Temperature T (°C) Figure 10. Plot showing tem- 25 30 35 40 perature versus sample depth (left) and diagrammatic col- umns showing the defi nitions Ts = 25 °C Tmax = 35 °C of variables in Equation 4 (right). Plot illustrates five- Upper crustal 1000 step graphical procedure to columns at : estimate maximum paleodepth

dT = ° (m) z for shallow samples, given h 25 C/km 70 Ma 0 Ma max dz z an upper bound of temperature 5 max T from thermochronology. = 200 m max 500 h = 500 m Adjustments from the modern 3 SAMPLE 70 = sample elevation and surface Elevation h 400 m 1 0 temperature (step 1) include h ΔT 4 r climate change (step 2), rock 2 =100 m uplift (steps 3 and 4), and the 0 effect of elevation on surface h temperature (step 5). See text r for discussion. dTs = – 8 °C/km dh h70 = ( h 0– h r ) + z max

16 Geological Society of America Bulletin, Month/Month 2010 Geological Society of America Bulletin, published online on 26 January 2011 as doi:10.1130/B30274.1

California River

2 m.y., L = 11 km. Because the total temperature ally compensated by the fl exurally rigid Colo- is negative). Hence, on the basis of the AHE ages variation is nearly 300 °C over this depth inter- rado Plateau (Lowry and Smith, 1995) and for and known surface temperatures, the sample re- val, the perturbation has a much smaller effect our purposes is small enough to be neglected, sided between 25 and 35 °C after 70 Ma, with on the geothermal gradient than shorter-term, but a number of authors have proposed varying no basis to prefer any particular value within the near-surface variations, and acts over a length amounts of late Cenozoic rock uplift of the pla- range. Similarly, there is no preference for any scale at least an order of magnitude greater than teau, which is taken into account by Equation 4. particular value for the decrease in MAT within any estimate of zmax. Hence, we approximate the For h = h70, substituting Equation 4 into Equa- the 3–8 °C range. effect of long-term climate change simply as a tion 3 yields Using these parameters, Equation 6 becomes static shift of the geotherm by ΔT, neglecting its Δ Δ Δ small, positive effect on dT/dz if T < 0, which Ts h70 = Th = 0 – T + (dTs/dh)(h0 – hr + zmax). (5) zmax = [(Tmax + T) – (25 ± 2 °C)]/ in any event is at best known to within ±10%. (17 ± 2 °C/km). (8)

This approximation is justifi ed by the fact that Substituting this expression for Ts h70 into Equa-

high-quality measurements of Ts in shallow tion 1 and solving for zmax yields The two parameters with standard deviations of boreholes (<200 m) throughout the southwest- ~10% do not introduce large errors into the esti- ⎡ ⎤ Δ ern United States exhibit systematic variation []+ Δ −+dTs ()− mate of zmax, but the sum of (Tmax + T) within TTTmax ⎢ h=00hhr ⎥ with elevation (Fig. 11), whereas dT/dz does not ⎣ dh ⎦ the limits for each parameter stated above var- zmax = . (6) (e.g., Sass et al., 1994). dT + dTs ies by nearly a factor of 2, ranging from 17 to As discussed already, we also neglect the ef- dz dh 32 °C. Using the best estimates of the other fect of climate change on lapse rate, yielding a two parameters in Equation 8, these limits on Δ Tertiary surface temperature Equation 6 is a refi nement of the customary (Tmax + T) yield a range of zmax of –470 to equation for estimating much larger paleodepths +412 m. Negative values are, of course, impos- Δ Ts (h) = Th = 0 – T + (dTs/dh)h. (3) on the basis of thermochronometric data sible because of the existence of the sample and Earth beneath it, but the upper limit suggests that

The elevation of Earth’s surface at the onset zmax = (Tmax – 20 °C)/(dT/dz), net erosion since the Cretaceous does not ex- of post–70 Ma erosion, h70, is simply ceed ~400 m, a fi gure less than half the 1000 m where the precise surface temperature, the ele- maximum erosion implied by the northward

h70 = (h0 – hr) + zmax, (4) vation dependence of surface temperature, rock projection of the Peach Springs–Truxton paleo- uplift, and climate change are neglected. canyon discussed previously (Fig. 8, measure-

where hr is rock uplift, defi ned as the net up- The right-hand side of Equation 6 contains ment 4), and 200 m less than the 600 m erosion ward displacement of bedrock relative to sea seven parameters, all of which can be estimated estimated at the beginning of this section on the level (Fig. 10). For the small amounts of total on the basis of independent measurements. The basis of thermochronology. erosion under consideration and the relatively greatest uncertainty is contained in the fi rst two The 400 m fi gure, however, requires a co- narrow aperture of canyon erosion (<20 km), terms in the numerator, the sum of the maxi- incidence of extreme values of residence tem- we assume that any isostatic rebound is region- mum sample temperature and climate change. perature and climate change. The probability The remaining four terms in the numerator is of net erosion exceeding a certain value can be simply the present surface temperature of the estimated using a Monte Carlo approach, as- 2000 sample, which is 25 ± 3 °C, neglecting rock up- signing equal probability to the occurrence of ° Δ Ts = –0.008h + 29 C lift (Fig. 10). Regression of the measured sur- Tmax and T per degree centigrade within their R 2 = 0.87 face temperatures from Sass et al. (1994) as a stated ranges. A value of zero net erosion cor- function of elevation (Fig. 11) yields responds to T = 28 °C (but in this case, only 1500 max the value of ΔT = –3 °C is possible), limiting

Ts(h) = (29 ± 2) °C + (–8 ± 1 °C/km)h. (7) the total possible range to 28–35 °C, or eight (m) h discrete values. For ΔT, there are six discrete 1000 The linear fi t is surprisingly good (R2 = 0.87) values between –3 °C and –8 °C. The multipli- considering that the thermal conductivities and cation rule yields 48 outcomes with a symmetric Elevation Elevation geothermal gradients in the data set both vary by distribution, of which 15 combinations result in 500 more than a factor of 2, and confi rms that MAT negative net erosion. The resulting distribution

is the primary control on shallow surface tem- indicates the highest probabilities for 0 < zmax perature. As is clear from Equation 6, adding < 235 m (36/48 cases or 75%), with only 1 of 0 positive rock uplift has the effect of decreasing 48 outcomes (2%) yielding the value of 412 m. 10 15 20 25 30 the estimate of zmax (n.b. that dTs/dh < 0). The Taking these outcomes as an approximation for Surface temperature Ts (°C) denominator is the geothermal gradient minus a Gaussian distribution, and ignoring (1) the

the absolute value of the surface temperature asymmetry introduced by negative values of zmax Figure 11. Plot of surface temperature ver- lapse rate, or (25 ± 2 °C/km) + (–8 ± 1 °C/km) = and (2) the second and third signifi cant fi gures, sus elevation in shallow boreholes through- 17 ± 2 °C/km. we fi nd an upper limit of zmax ~400 m at two out the southwestern United States. Plot was The lower extreme for the residence tem- standard deviations. constructed by extrapolating temperature- perature in western Grand Canyon is simply the By contrast, a depth of zmax = 1500 m for most depth curves (fi gs. A2 through A8 in Sass present surface temperature, or 25 °C, assuming of the post–70 Ma period, assuming ΔT = –3 °C et al., 1994) to Earth’s surface. Regression signifi cant lowering of the average elevation of and no rock uplift, yields Tmax > 54 °C, more equation assumes h in meters. the plateau did not occur (i.e., the case where hr than 19 °C too warm, and predicts AHE ages

Geological Society of America Bulletin, Month/Month 2010 17 Geological Society of America Bulletin, published online on 26 January 2011 as doi:10.1130/B30274.1

Wernicke of less than 20 Ma, at least 40 m.y. too young source for the orthoquartzite clasts in the Sespe because it would require ~2° of tilting of the (Fig. 6). Accounting for the Hualapai Plateau Formation, as originally contemplated (but re- Coco nino terrace (>1100 m in 40 km requires a paleocanyons and other data, a relatively mod- jected) by Howard (1996, 2000). tilt of 1.6°, plus the northward slope of the chan- est post–6 Ma erosion of 700 m, coupled with nel). Allowing the structurally low channel an 250 m of rock uplift was proposed by Karlstrom Implications for the Outlet of the Hualapai outlet through the Grand Wash Trough (hypoth- et al. (2008). This case, with ΔT = –3 °C, yields Plateau Paleocanyons esis 2) thus better explains the facts in western

Tmax > 43 °C, still more than 8 °C too warm, and Grand Canyon, if not the Muddy Creek problem. predicts AHE ages of less than 45 Ma, at least Pursuant to the Sespe question, the most Given these diffi culties, it is appropriate to re- 15 m.y. too young. interesting implication of limiting Cenozoic examine evidence bearing on whether the paleo- A fi ve-step graphical procedure generally incision at Diamond Creek to <400 m is that canyons lay upstream from the Lower Tertiary applicable to problems of this type allows di- it requires a local base level for the Peach fl uviolacustrine deposits in Utah, which was an rect visualization of the relationship between Springs–Truxton paleocanyon at least 300 m, important motivation for the Young-McKee hy- the primary measurements and maximum ero- and perhaps as much as 600 m, lower than its pothesis. It presumed (1) synchronism of Rim sion (Fig. 10). In a space of elevation h ver- projected position at Grand Canyon (measure- gravel channel aggradation and fl uviolacustrine sus temperature T, we plot the modern sample ment 4, Fig. 8), if I have interpreted the error deposition of the Claron Formation and related elevation h0 on the modern curve of dTs /dh budget for zmax correctly. deposits in southern Utah, and (2) a hydrologi- (Fig. 10, step 1), and then shift the curve up- Any model that invokes post–6 Ma inci- cal connection between the Rim gravels and temperature by the estimate of the climate sion of westernmost Grand Canyon of more various gravel bodies north of Grand Can- Δ change T (step 2). We locate the point (Tmax, than 700–1000 m precludes simply routing the yon (Elston and Young, 1991). As to point 1, h0) (step 3) and shift it downward by the paleo canyon NW down the westernmost seg- whether the Rim gravels are uppermost Eocene amount of rock uplift hr (step 4). The maxi- ment of Grand Canyon in Paleogene time, be- instead of lower Eocene, it would be consistent mum burial depth may then be determined by cause it would require the river to fl ow uphill with the hypothesis, within permissible limits projecting a line with slope of dT/dz upward between the Diamond Creek area and the Grand on the age of the Claron Formation. An impor- from the point obtained in step 4 to its inter- Wash Cliffs. Young incision models thus present tant test as regards point 2 is comparison of the section with the climate-adjusted curve for a sort of “River Styx paradox” for the outlet of provenance of the Rim gravels with (a) the ex-

Ts(h) obtained in step 2, which yields an es- the paleocanyon relative to the modern topogra- posures of “Canaan Peak–type” gravels north of timate for h70, and therefore zmax (step 5). This phy towering around it. This paradox led to the the river, (b) Cretaceous–Paleogene strata ex- Δ approach, for Tmax = 35 °C, mid-range T = suggestion (hypothesis 1, Fig. 8) that the Peach posed in southern Utah, and (c) Paleogene strata 5 °C, and 100 m of rock uplift yields a maxi- Springs–Truxton paleocanyon escaped north- in coastal California. mum erosion of 200 m (Fig. 10), which can be ward across the north rim of Grand Canyon to- As noted already, the Canaan Peak–type analytically verifi ed by using Equation 6. The ward southern Utah (Young and McKee, 1978; gravels north of the canyon are undated, and utility of the plot is that it allows visualization Young, 1979; Elston and Young, 1991). Accord- therefore could entirely postdate a hypotheti- of how uncertainties in dT/dz, dTs /dh, hr, Tmax, ing to this hypothesis, the ancient river need not cal northward paleoslope connecting the Rim Δ and T affect the estimate of zmax, and in the have had a trajectory that climbed from 1050– gravels with Eocene deposits in southern Utah. present case, shows that it is diffi cult to honor 1350 m at Diamond Creek to nearly 1900 m im- Thus, while not precluding the Young-McKee all the constraints with more than a few hun- mediately to the north, so as to clear the Kaibab hypothesis, the transport direction and prov- dred meters of erosion. escarpment at the southern end of the Shivwits enance of these gravels, at least as now under- Regardless of the details of these estimates, Plateau (Fig. 8). North of Diamond Creek, stood, do little to motivate it. with only a few hundred meters of erosion, Grand Canyon runs east of the Shivwits Plateau, The source of the Rim gravels, which con- western Grand Canyon landscape in the Dia- in alignment with the northerly trend of the tain abundant Cenomanian–Maastrichtian vol- mond Creek area is not dramatically different Peach Springs–Truxton paleocanyon (Fig. 2). canic clasts, could also have been a source of today than it was at 70 Ma, modifi ed primarily Thus, the paleocanyon, assuming a base level at Campanian zircons in the pre-Maastrichtian by aggradation of the Rim gravels and overly- Diamond Creek of ~1350 m, could climb north- section in southern Utah (e.g., Dickinson and ing strata, and offset as much as a few hundred ward an additional 350 m to clear the north rim Gehrels, 2008), consistent with the Young- meters along the Hurricane fault zone. This at 1700 m elevation, over a distance of at least McKee hypothesis. However, the overall pat- estimate of erosion is surprising only in that it 40 km to the north, where the river bends back to tern of (1) progressively more restricted source extends the level of pre–Rim gravel incision, the east and the Kaibab escarpment trends E-W regions in the Pine Hollow, (2) the lower already known from the Hualapai Plateau to (Young, 2001a; Figs. 2 and 8). Hypothesis 1 for Claron’s southward transport direction, and be at 1050–1350 m elevation just 8 km south the channel’s outlet thus requires extrapolating (3) the lack of Cretaceous zircon altogether in of the canyon, down to a level of between 400 the SW tilt of the paleocanyon on the Hualapai the Claron, does not support a connection be- and 800 m elevation within the canyon itself Plateau northward to affect most of the width of tween the Claron basin and far-traveled detrital (Fig. 8). Given that the Colorado River in the the Coconino terrace (Young, 2001a). input from a Rim gravel source, in either early Diamond Creek area is locally incised as much If, however, base level for the paleocanyon at or late Eocene time. If anything, the data are as 275 m below the basal Cambrian unconfor- Diamond Creek was at just 600 m elevation, it suggestive of a barrier between any Rim gravel mity (Billingsley et al., 1999), the level of ero- requires that the paleocanyon, rather than gen- source and southern Utah from Maastrich- sion through much of Tertiary time would most tly rising toward Grand Canyon, instead sloped tian time onward. Thus, as is the case with the likely have been near the level of the Tapeats northward at 3°–4° between extant Rim gravel Canaan Peak–type gravels, the provenance Sandstone or below. This in turn opens up the deposits and the bottom of the canyon (hypothe- and paleohydrology of Paleogene sandstone possibility that a deep, Paleocene–Miocene sis 2, Fig. 8). For this hypothesis, the task of sur- and conglomerate in southern Utah offer little western Grand Canyon was indeed a potential mounting the north rim becomes quite diffi cult, support for a hydrological connection between

18 Geological Society of America Bulletin, Month/Month 2010 Geological Society of America Bulletin, published online on 26 January 2011 as doi:10.1130/B30274.1

California River highlands SW of Grand Canyon and the plateau yons of the Hualapai Plateau and western Grand at least an order of magnitude higher than the in Paleogene time. Canyon (and presumably also eastern Grand observed rates (Dallegge et al., 2001). In recognition of Goldstrand’s (1994) con- Canyon, as elaborated in the following) were As also pointed out by Young (2008), mod- clusion of western and northern sources for the hydrologically connected to the coast during ern drainages feeding the Colorado River below Claron, Young (2001b) proposed a modifi ca- Paleogene time, how could such a deep, long- Lees Ferry, which constitute 22% of its total tion of the Young-McKee hypothesis wherein lived gorge leave no evidence of its existence drainage area above Lake Mead, contribute only Rim gravel detritus accumulated in a closed in the Muddy Creek Formation or its substrate 4% of the total discharge of 500 m3/s. The Little basin in northern Arizona in early Eocene time. during Miocene time? Even more enigmatically, Colorado River drainage, which comprises This basin may have been physically continu- why was such a deep gorge carved in Campan- 19% of the total drainage area upstream from ous with the Claron basin, but did not share its ian and perhaps early Maastrichtian time? Such eastern Grand Canyon gauge at Bright Angel northern and western source regions. However, an event would presumably have required the Creek, contributes <1% to the total discharge. thermochronological data presented previously combination of a long, fi rst-order trunk stream, However, the contributions of these drainages to favoring hypothesis 2 over hypothesis 1 in Fig- coupled with regional, kilometer-scale tectonic the sediment load of the Colorado are at pres- ure 8 are diffi cult to reconcile with the steep uplift (i.e., base-level fall) to drive incision of ent substantial. Construction of Glen Canyon northward climb of the drainage system out of the enormous canyon. dam near Lees Ferry cut off the sediment load the Diamond Creek area of Grand Canyon, as above the dam, such that the postdam drainage, would be required by this model. A Possible Resolution to the as regards sediment delivery, is similar to the The provenance of Paleogene detritus in Muddy Creek Problem confi guration proposed here for Muddy Creek coastal California, on the other hand, is con- The problem of the absence of mature fl u- time. The dam reduced the sediment load of sistent with a hydrological connection between vial detritus in the Muddy Creek Formation the Colorado in Grand Canyon by 83%, from Rim gravel canyons in the Grand Canyon area was recently addressed by Young (2008). He 83,000 to 14,000 Mg/yr (Topping et al., 2000). and the coastal basins. The absence of Creta- proposed that because Grand Wash Trough and Thus, even though tributaries to the Colorado ceous arc detritus in the Claron basin and its basins to the west were occupied by a large below the dam contribute 4% of the discharge, abundance in both the Rim gravels and coastal lake for much of their existence (Miocene Lake they contributed 17% of the predam sediment basins suggest a convex-north “U”-shaped drain- Hualapai of Spencer et al., 2008a), and that the load. This raises the question of whether the age system where NE-transported Rim gravels now-dissected lake sediments lie at elevations Grand Wash Trough–Muddy Creek basin could south of Grand Canyon were directed into west- ranging from 400 to 900 m (up to 500 m above have accommodated the sediment load from a ern Grand Canyon and then southward to the the modern river grade; Fig. 3), an arm of the drainage system as large as Grand Canyon and coast (Fig. 9B), consistent with hypothesis 2 in lake probably extended eastward up through an its tributaries below Glen Canyon. Figure 8. The drainage reversal history in the actively incising western Grand Canyon in Mio- For a closed basin of area A, assuming uni- Grand Canyon region would thus contrast with cene time. Because of this, any river sediment form distribution of sediment, the annual aggra- the history of the Rim gravels in the Fort Apache supplied from the integrating drainage system dation would be region (Fig. 1), where the drainage system in would be trapped in a delta many tens of kilome- late Eocene time transported detritus from the ters upstream, and therefore the only source for h = m/ρA, Transition Zone ENE onto the plateau in west- coarse detrital deposits in Grand Wash Trough central New Mexico (e.g., Potochnik, 2001). would be local. where h is the thickness, m is the mass (an- The persistence of Mojave/Mogollon high- The existence of a deep western Grand Can- nual sediment load), and ρ is the density of land detrital input to the Uinta basin along the yon throughout Cenozoic time, as suggested the added sediment. For a closed Grand Wash northern margin of the plateau into early Eocene here, requires the existence of an even larger Trough of dimensions 50 km × 20 km, assum- time (Davis et al., 2010) is diffi cult to reconcile Lake Hualapai than envisaged by Young (2008), ing today’s postdam sediment load and sedi- with this conclusion, because it seemingly re- who depicted the headwaters of the lake in ment density of 2000 kg/m3, it would aggrade quires a hydrological connection between the middle to late Miocene time to lie west of the at a rate of 7 mm/yr, i.e., two orders of magni- two regions at a time when data from the Grand Kaibab arch, cut along escarpments in Upper tude greater than the observed Miocene rate of Canyon region and southern Utah suggest a Paleozoic strata. Even assuming erosion of as ~0.07 mm/yr, and would result in aggradation reversal in paleofl ow direction had already oc- much as 400 m in western Grand Canyon and between 13 and 6 Ma of 49 km of sediment. curred. This paradox is addressed in the context placing it all post–6 Ma, the 900 m elevation Even distributing the sediment over the entire of the paleohydrological model proposed next. of the Hualapai Limestone fi ll surface in Grand area of Lake Hualapai, roughly four times larger Wash Trough implies that the lake would have than Grand Wash Trough (Spencer et al., 2008a), Alternative Hypothesis been at least 100 m deep at Diamond Creek, it would still require an average of >10 km of and may have backed up as far as eastern Grand sediment deposited over a very large area. To my knowledge, all contemporary hypoth- Canyon, depending on the details of the late his- Clearly, Grand Canyon could have existed eses for the evolution of Grand Canyon, even tory of erosion and tectonism. However, is such before 6 Ma only if rainfall was insuffi cient to those that suggest substantial carving of por- a large hypothetical drainage area compatible support streams that carry large sediment loads. tions of Grand Canyon well before 6 Ma (e.g., with a lack of fl uvial detritus in Grand Wash Paleobotanical data and climate modeling sug- Potochnik, 2001; Scarborough, 2001; Flowers Trough? The same point applies to drainage gest the region was not only much drier than et al., 2008; Hill and Ranney, 2008; Young, integration of the upper Colorado River basin at present, but also probably lacked intense 2008), do not explicitly challenge some form during Bidahochi deposition. Assuming modern summer rainfall (Young, 2008, and references of piracy or spillover across the Kaibab arch– sediment loads of the upper Colorado River ba- therein). Hence, in contrast to the modern sys- Coconino terrace region, primarily because of sin are applicable to the Miocene, aggradation tem, where vegetation and storminess are opti- the Muddy Creek problem. If the Paleogene can- rates in the Bidahochi basin are predicted to be mum to produce large annual sediment loads

Geological Society of America Bulletin, Month/Month 2010 19 Geological Society of America Bulletin, published online on 26 January 2011 as doi:10.1130/B30274.1

Wernicke in relatively low-discharge streams, arid land- only if the modern storm-driven sediment loads cusing groundwater discharge that was just suf- scapes with rare intense storms would be ex- are applicable to Miocene time. fi cient to maintain permanent lakes, which was pected to have a greatly reduced discharge and Even though rainfall patterns may have been frustrated by evaporation in the relatively small sediment fl ux. insuffi cient to support sedimentary transport up- surrounding drainages. This interpretation obvi- For example, large drainage areas issuing stream from the Grand Wash Cliffs, there still ates the need for either subterranean hydrologic southward from the Oman Mountains (1000– may have been suffi cient rainfall to support discharge or headward erosion as agents to in- 2000 m in elevation) on the northeastern Arabian groundwater discharge from Grand Canyon to cise Grand Canyon. Peninsula are a typical example of an arid land- gradually fi ll and expand Lake Hualapai (e.g., The last element of the Muddy Creek prob- scape that lacks summer rainfall, and thus may Hunt, 1969). In this scenario, the lake may lem to resolve is identifying the course of the provide a modern analogy for drainage systems have accumulated a small amount of both fi ne river west of Grand Canyon prior to the forma- in the southwestern United States in Miocene sediment and dissolved load from the carbon- tion of Grand Wash Trough, where, according time. Sedimentation on the lowland plains im- ate uplands (Young, 2008). I suggest that this to hypothesis 2 (Fig. 8), it must have fl owed mediately to the south of the range is character- scenario would be favored by having the entire prior to Muddy Creek deposition. Wheeler ized by meter-scale aggradational events roughly modern Grand Canyon below Lees Ferry, not Ridge and other tilted fault blocks in the Grand every 100,000 yr since 0.5 Ma (Blechschmidt yet integrated with its Rocky Mountain sources, Wash Trough form a continuous belt of expo- et al., 2009). Average deposition rates are thus draining into Grand Wash Trough and environs, sures of tilted Cambrian through Permian strata on the order of tens of meters per million years, because it would be more effective in promoting that, when palinspastically restored against the comparable to the Muddy Creek and Bidahochi the development of large lakes than would the Grand Wash Cliffs (Brady et al., 2000), would Formations, and include very long stretches of relatively small precursor drainage suggested by seemingly lie athwart the western terminus time without aggradation. Correlation between Young (2008). of Grand Canyon. However, previous work- these depositional events and precipitation events This hypothesis is supported by the fact that ers have noted the presence of a paleocanyon in caves has been interpreted to refl ect infrequent deposition of the Hualapai Limestone in the in the southern part of Wheeler Ridge near encroachment of monsoonal moisture from the Lake Mead region contrasts strongly with co- Sandy Point (Figs. 2 and 12A), where there east onto the northern Arabian Peninsula during eval deposits in adjacent basins, in which evapo- is a 3.5-km-wide gap in exposures of moder- periods of postglacial warming (Blechschmidt rites predominate over limestone, refl ecting a ately to steeply dipping Paleozoic section and et al., 2009). Winter rainfall in the Oman Moun- climate that was generally too arid to form per- underlying basement rocks (Longwell, 1936; tains is comparable to that of ranges in the manent lakes (Hunt, 1969). Faulds et al. (2001, Lucchitta, 1966; Wallace et al., 2005). The southwest United States, indicating that summer p. 87) suggested that “much of the fresh water canyon fi ll is gently east-dipping (0°–10°), storminess is essential for the generation of large was probably derived from the western part very coarse, poorly sorted conglomerate and annual sediment loads in arid or semiarid drain- of the Colorado Plateau through springs issu- interstratifi ed rock-avalanche deposits derived ages. Without periodic encroachment of mon- ing from Paleozoic limestone and/or a system from the paleocanyon walls and from Protero- soon moisture, there would have been little or no of headwardly eroding streams that eventually zoic basement in the South Virgin Mountains aggradation on the southern plains since 0.5 Ma. evolved into the Colorado River.” It is proposed (Fig. 12A). The provenance and high-energy Thus, the existence of the Grand Canyon during here that the contrast with the surrounding ba- depositional facies indicate generally eastward Muddy Creek time is precluded by the slow ag- sins instead resulted simply from having Grand transport of this detritus from the South Virgin gradation rate of the Muddy Creek Formation Canyon, already formed, as its headwaters, fo- Mountains into the Grand Wash Trough, with

Figure 12. (A) Geologic map of Wheeler Ridge paleocanyon, simplifi ed from Wallace et al. 4 5

A 2 (2005); lines with tick marks N Rock avalanche Th and numbers show strike and deposits in DM dip of bedding. (B) Cross sec- W h e e l e r R i d g e paleocanyon fill Tm 46 41 C tion through western Grand DM P 33 Canyon for comparison, using by faulting C 63 33 geology from Wenrich et al. exposure Xg of Tc

(1996). See Figure 2 for lo- 114°05’00”W Limit cation of map and section. 1 km PRE-MUDDY CREEK NONCONFORMITY Xg—Protero zoic crystalline 36°02’30”N basement; C—Cambrian strata; NNE SSW B Sanup Plateau Grand Canyon P Hualapai Plateau DM—Devonian–Mississippian 1500 P strata; P—Pennsylvanian strata; DM Tc—Tertiary conglomerate and 1000 DM intercalated ash beds with sani- 500 C dine yielding an Ar-Ar age of Elevation (m) Elevation Xg 15.3 ± 0.1 Ma; Tm—Muddy 0 NO VERTICAL EXAGGERATION Creek Formation (sedimentary rocks of Grand Wash Trough); Th—Hualapai Limestone.

20 Geological Society of America Bulletin, Month/Month 2010 Geological Society of America Bulletin, published online on 26 January 2011 as doi:10.1130/B30274.1

California River the paleocanyon representing a trunk stream Grand Wash Trough, after most faulting and tilt- California River for the distribution of sediment in Grand Wash ing had ceased. However, that erosional event The thermochronological, paleoaltimetric, Trough, mainly by infrequent rock-avalanche was not necessarily responsible for all, or even and provenance data suggest that an E-fl owing and debris-fl ow events (Lucchitta, 1966, 1979). a signifi cant fraction of, the erosion that created paleocanyon was cut to a depth of 1500 m in Detailed mapping of the paleocanyon walls the paleocanyon. The normal faults truncated by Campanian time, to a level near the present indicates that it probably had a NE trend as it the unconformity have displacements of only a erosion surface in western Grand Canyon, and was being fi lled, and that the unconformity at few hundred meters at most, and therefore mod- to the level of Lower Mesozoic strata (at river the base of the fi ll overlaps normal faults within est post-tilt erosion would have eliminated any level) in eastern Grand Canyon (Figs. 13A and the Paleozoic section (Wallace et al., 2005). At tectonically generated relief prior to aggradation 13B). Near 20 Ma, prior to the extensional event the southern end of the paleocanyon, in an iso- of the fi ll over the paleocanyon wall, long after that formed Grand Wash Trough, the erosion lated fault block just west of the main ridge, a most tilting and faulting within Wheeler Ridge surface in eastern Grand Canyon was lowered to pre–Muddy Creek, Paleozoic-clast conglom- had ceased. within a few hundred meters of its present posi- erate unit and overlying ash deposits dated at The hypothesis that Wheeler Ridge contains tion without a major change in relief (Figs. 13D 15.3 ± 0.1 Ma were mapped by Wallace et al. a fragment of Grand Canyon is testable using and 13E). Post–20 Ma erosion throughout the (2005) as resting nonconformably on Proterozoic a comparison of a down-plunge view of the canyon has not exceeded a few hundred meters, basement, with two measured dips in the unit Wheeler Ridge paleocanyon (essentially a map with perhaps slightly more erosion possible in both at 33°E (Fig. 12A). Attitudes measured in a view, Fig. 12A) and a vertical north-south cross eastern Grand Canyon (Fig. 4). transect through the Paleozoic section of the main section of Grand Canyon just upstream from its The existence of a deep canyon since the Cre- ridge immediately to the east by Wallace et al. intersection with the Grand Wash Cliffs (Figs. 2 taceous raises the questions of why it was cut and (2005) are (from W to E, going up section) 24°, and 12B). The key question is whether the which way the river was fl owing. Based on prov- 41°, 24°, 21°, 41°, 47°, and 73° (mean = 35°), geom etry of the paleocanyon is compatible with enance data on either side of the orogen, it is clear and so, on average, particularly in the Lower the dimensions of western Grand Canyon. The that the active California arc was feeding large Paleo zoic units closest to the fault block, the con- comparison shows that the horizontal N-S sepa- volumes of detritus through a major drainage glomerate appears to have been deposited prior ration between any two formations in modern directly onto the Colorado Plateau, through at to most of the tilting of the ridge, and is older Grand Canyon is about the same as the horizon- least the end of Campanian time (71 Ma). This than the more gently dipping bulk of the canyon tal separation between the same formations on marked the end of a long period in the Late fi ll unit, which to the north contains an ash bed either side of the Wheeler Ridge paleocanyon, Cretaceous when the Cordilleran mountain belt dated at 10.94 ± 0.03 Ma (Wallace et al., 2005). along sections where the modern canyon is most was relatively narrow, and the region east of the Wallace et al. (2005) entertained two hy- narrow. For example, in both sections, the width Sevier belt/miogeoclinal hinge zone (north) or potheses for the origin of the paleocanyon. In of the canyon at the position of the Redwall the arc (south) lay near sea level as the Western the fi rst, it is analogous to one of the Paleogene Limestone is 3–4 km. A deep, pretilt “notch” cut Interior Seaway withdrew to the east (Fig. 9A; canyons preserved on the Hualapai Plateau, and into the top of the future Wheeler Ridge fault e.g., Dickinson et al., 1988). The carving of the was later fi lled with Neogene rather than Paleo- block would form a natural topographic depres- paleo canyon took place during Campanian and gene detritus. In the second, the canyon was sion even after signifi cant rotation of crustal not carved until Neogene time, in response to blocks. Therefore, during and after tilting, the the formation of Grand Wash Trough and the ancient channel would likely have been ex- tilting of fault blocks. They preferred the sec- ploited as a local topographic low that focused Figure 13. Diagrammatic cross sections of ond hypothesis, on the basis that the ancient fi ll drainage eastward through the paleocanyon and the six areas highlighted in yellow in Fig- along the canyon wall truncates Miocene nor- into the Grand Wash Trough. ure 9. Wavy lines at the top of each diagram mal faults. However, the observation that pre- or show the elevations of river grades (bottoms early tilt gravel lies unconformably on Protero- Late Cretaceous to Quaternary of V-shaped depressions, connected by ar- zoic basement in the vicinity of the canyon fa- Paleohydrology of the Colorado River rows) and surrounding uplands (to the left vors the existence of a deeply incised canyon Basin from the Glen Canyon Area and right of the depressions), with key forma- prior to extensional tectonism. to the Coast tions in depositional basins labeled in italics. I propose that the Wheeler Ridge paleocan- Geologic units are Proterozoic crystalline yon is a tilted fragment of Grand Canyon. This The possible resolution of the Muddy and overlying Proterozoic stratifi ed rocks hypothesis is a composite of the two hypotheses Creek problem and consideration of thermo- (brown), Paleozoic strata (light blue), Trias- suggested by Wallace et al. (2005), wherein the chronological, paleoaltimetric, and sedimen- sic through Lower Cretaceous strata (forest steep walls of Grand Canyon, originally formed tary-provenance data suggest a three-phase green), Upper Cretaceous strata (chartreuse), in Late Cretaceous time, were rotated eastward paleohydrological reconstruction for the south- Paleogene strata (gold), Upper Oligocene during rifting as a part of the Wheeler Ridge western United States, involving two major through mid-Miocene strata (orange), and block, except that the canyon had westward drainage transitions, one near the Cretaceous- mid- to late Miocene strata (yellow). Also paleo fl ow immediately prior to rifting rather Tertiary boundary and another in the late Mio- shown are ophiolitic basement of the forearc than eastward, and hence was originally down- cene (Fig. 13). The transitions defi ne three terrain (olive green), metamorphosed sub- stream from the Peach Springs–Truxton paleo- contrasting drainage networks. Using a conven- duction complex rocks (gray and olive), and canyon as described previously, rather than a tion for naming a river after the state contain- Mesozoic arc intrusive rocks (pink). Sediment co-tributary draining to the NE. ing its headwaters, I will refer to the fi rst system load in rivers is indicated qualitatively as high The erosion surface that truncates normal as the California River and to the second as (three arrows), moderate (two arrows), and faults within Wheeler Ridge refl ects erosion the Ari zona River, the third of course being the low (one arrow). Horizontal gray lines show that occurred late in the history of formation of modern Colorado River. elevation above sea level.

Geological Society of America Bulletin, Month/Month 2010 21 Geological Society of America Bulletin, published online on 26 January 2011 as doi:10.1130/B30274.1

Wernicke

SW Grand Canyon Glen NE Los Angeles Needles Kingman West East Canyon 2000 m 1000 m A 80 Ma Wahweap Sea Sea level level ARC SUPPLY CALIFORNIA CORDILLERAN AXIS RIVER GRAND CANYON INCISION UPLIFT TRENCH/FOREARC HINGE B 70 Ma 2000 m 1000 m Kaiparowits Sea Tuna Sea level Canyon level ARC SUPPLY

TECTONIC JUXTAPOSITION

Pelona Schist Claron C 55 Ma 2000 m ARC 1000 m Simi CUTOFF EXOTIC 0 m DETRITUS APPEARS

RUNYON CANYON ASYMMETRIC SURFACE ARIZONA RIVER DRAINAGE DIVIDE Rim gravels Chuska 2000 m D 30 Ma 1000 m Sespe 0 m EXOTIC DETRITUS SUPPLY

TECTONIC DERANGEMENT EQUILIBRIUM OF ARIZONA RIVER DRAINAGE UNROOFING Bidahochi Copperpp Horse 2000 m E 16 Ma Basin Springpg 1000 m Topanga 0 m EXOTIC DETRITUS CUTOFF

SPILLOVER Muddy 2000 m Creek 1000 m F 6 Ma Bouse Sea 0 m level Puente

MODERN COLORADO RIVER CONTINENTAL DIVIDE 2000 m SAdflSan Andreas fault G Present 1000 m 0 m

Figure 13.

22 Geological Society of America Bulletin, Month/Month 2010 Geological Society of America Bulletin, published online on 26 January 2011 as doi:10.1130/B30274.1

California River perhaps the early part of Maastrichtian time periods of sediment transport transverse (SE) to composite of Mojave/Mogollon highlands prov- (ca. 80–70 Ma, given the uncertainties on the the Sevier front (Fig. 9A; Lawton et al., 2003), enance along the coast in Paleocene time (e.g., AFT and AHE ages), by a hydrologically im- and probably existed farther SE in Arizona, per- Simi Conglomerate at >56 Ma) coeval with its portant river with extensive headwaters to the haps along an axis parallel to the late Eocene disappearance from deposits in southern Utah SW (Fig. 9A). This period, therefore, marked drainage system between the Fort Apache re- ( Paleocene–Eocene Pine Hollow Formation). the beginning of a wave-like expansion of both gion and the Baca basin (Fig. 9; Potochnik, Hence, at least in this region, depocenters on topographic uplift and erosional unroofi ng into 2001). AHE cooling ages and stratigraphic data both sides of the orogen indicate expansion of the Cordilleran foreland (e.g., Flowers et al., along the Arizona homocline between the Fort the coastal drainage networks and contraction 2008; Spencer et al., 2008b), and was coeval Apache region and the transverse drainages in of the interior networks (Figs. 9B and 13C). with Campanian deformation that disrupted southern Utah do not suggested the presence of Drainage of the interior toward the coast also drainage patterns to the east in the Rocky Moun- any deep Cretaceous canyons other than Grand affected portions of southern and eastern tains (Cather, 2004). In the earliest part of the ex- Canyon (Flowers et al., 2008). Arizona in early Tertiary time (e.g., Kies and pansion near the beginning of Campanian time, To the NE of the arc from 93 to ca. 75 Ma, Abbott, 1983; Howard, 2000). Farther north Cordilleran highlands were still focused on the the SW margin of the Colorado Plateau was in central Nevada, studies of Paleogene drain- arc, which was an area of active volcanism, high thus a low-relief aggradational plane that lay age patterns suggest a topographic divide near relief, and rapid erosion (Figs. 9A and 13A). By near sea level, accumulating at least 1500 m longi tude 116°W (Henry, 2008; Fig. 9B). early Maastrichtian time, the expansion resulted of SW-derived sediment (Fig. 13A) that was The observation that the Paleogene Colton in kilometer-scale elevation and relief in the subsequently stripped away in Campanian Formation contains a similar arc-dominated southwestern part of the plateau, while areas to through early Eocene time (Fig. 13B; Flowers detrital zircon signature as the Campanian the NE were still low enough to trap sediment et al., 2008). Thus, during the Campanian, a Kaiparowits Formation (ultimately originating eroding from the arc far to the west. The supply NW-trending hinge zone developed between in the Mojave/Mogollon highlands) is seem- of Campanian arc detritus was eventually cut off the incising eastern Grand Canyon and aggrad- ingly in confl ict with the Maastrichtian cutoff from the foreland as a result of this process, as ing Kaiparowits Plateau region, near the present of arc-derived material in southern Utah, and recorded in the Canaan Peak Formation, which position of Lees Ferry (Fig. 13B). To the SW of with “hypothesis 2” in Figure 8, which implies was still receiving detritus from the eastern the arc, Maastrichtian basins in coastal Califor- that the drainage reversal in Grand Canyon had Mojave region after incision, but not from the nia record rapid, proximal deposition of detritus already occurred by early Eocene time. If the Campanian arc (at a stage intermediate between from the eastern arc terrain and its metamorphic California River persisted into early Eocene 70 and 55 Ma; Figs. 13B and 13C). The paleo- framework rocks, in response to tectonic events time (Davis et al., 2010), it would have to have canyon was thus incised by the California River, that formed the continental borderland province been hydrologically isolated from the Claron and was a major conduit delivering detritus from (Figs. 9B and 13B). basin. Alternatively, the Colton Formation may the topographic crest of the Cordillera in Califor- have been sourced in more proximal Laramide nia northeastward to the cratonic foreland. Arizona River uplands that were cut in Kaiparowits-equivalent The paleocanyon had the same approximate The cutoff of arc detritus in the southern Utah strata (e.g., along the fl anks of the Monument depth and position as modern Grand Canyon, basins during Maastrichtian time and syntec- upwarp, Fig. 9B). if not the precise level of erosion. The term tonic sedimentation during Paleocene–Eocene Assuming that the cutoff of arc detritus in “Grand Canyon,” in current usage, has been re- deposition of the Pine Hollow Formation indi- southern Utah signals the timing of drainage stricted by most authors to the modern one, go- cate that the main phase of Laramide deforma- reversal and that “hypothesis 2” is correct, I ing back to perhaps 6 Ma (e.g., Karlstrom et al., tion in southern Utah is distinctly younger than infer that during early Tertiary time, the in- 2008). This usage carries with it the genetic incision of Grand Canyon, and that it was coeval terior Laramide basins were separated from assumption that the canyon is young, having with major drainage reorganization. A compila- steep, SW-draining headwaters in the Grand attained its current morphology only over the tion of stratigraphic constraints on the forma- Canyon region by an asymmetric, NW-trending last 6 m.y. by the coalescence of a system of tion of major Laramide uplifts in the Rocky drainage divide located in what is now the Lee precursor drainages of markedly different depth Mountains indicates that after the onset of broad Ferry–Glen Canyon area (Figs. 9B, 13C, 13D, and geometry than the modern canyon (e.g., warping in Campanian time (based on the devel- and 13E). The primary motivation for the posi- Hill and Ranney, 2008), making it inappropri- opment of centripetal isopach patterns in Cre- tion and asymmetry of the divide is to provide a ate to apply the term “Grand Canyon” to any taceous strata), the main phase of deformation mechanism for subsequent drainage integration. feature older than 6 Ma. In the hypothetical (including steep local relief and stratal rotations) The geomorphology of this divide, if not the instance that the modern canyon eroded down- began in late Maastrichtian or Paleocene time precise tectonic setting, is envisaged to be simi- ward an additional kilometer but maintained its (e.g., Dickinson et al., 1988; Cather, 2004). This lar to the divide along the headwaters of the La modern position and depth, we would still refer general history is supported by fi ssion-track Paz River in South America, a component of to it as Grand Canyon, because a canyon is a studies, which generally yield Maastrichtian and the retroarc Amazon River drainage basin. The topographic feature defi ned by its morphology, younger (post–70 Ma) AFT ages from samples La Paz River near the city of La Paz, Bolivia, is not by its erosion level, age, or relationship to beneath fossil pre-Laramide PRZs (Kelley and unusual in comparison to most rivers that drain surrounding drainage networks. Therefore, if Chapin, 2004), although some older ages (up to the eastern side of the Andes in that it is eroding the hypothesis presented here is correct, then it 74 Ma) may refl ect cooling during Campanian headward into the Altiplano (Fig. 14). In con- would be most appropriate to refer to the Creta- unroofi ng (Cather, 2004). trast, other drainages along orogenic strike have ceous paleocanyon cut by the California River At a position along orogenic strike corre- headwaters in the much higher topography along as “Grand Canyon” (Fig. 13). sponding to Grand Canyon (transect highlighted the east fl ank of the Eastern Cordillera. The steep Other major conduits clearly existed during in yellow in Fig. 9 and depicted in Fig. 13), a gradient of the La Paz River contrasts with the this time in central Utah, as demonstrated by key observation is the appearance of detritus gentle backslope of the top of the plateau toward

Geological Society of America Bulletin, Month/Month 2010 23 Geological Society of America Bulletin, published online on 26 January 2011 as doi:10.1130/B30274.1

Wernicke

originated in eastern Grand Canyon and was fed by Howard’s (2000) orogen-parallel Gila- and L a k e T i t i c a c a (3812 m) Amargosa-Colorado paleorivers (Fig. 9B). The existence of an Amargosa-Colorado paleoriver Gentle, internal drainage to lakes and playas on Altiplano is supported by episodes of fl uvial aggradation from early Oligocene through middle Miocene time documented in the Death Valley–Amargosa D I V I D E (3900–4000 m) Desert region, Titus Canyon, Panuga, and Eagle D R A I N A G E Mountain Formations, which consistently exhibit southeasterly to southerly transport (Reynolds, A S Y M M E T R I C A L 1969; Snow and Lux, 1999; Niemi et al., 2001; Renik et al., 2008). R i v e r The model illustrated in Figures 9 and 13 is broadly similar to earlier models for the Steep, external drainage to Atlantic Ocean Cretaceous–Paleogene evolution of Grand Can- P a z yon, based on an analogy between Grand Canyon and the evolution of the Fort Apache region. L a These models include the elements of partial 1 km incision of Grand Canyon during the Laramide orogeny by NE-fl owing rivers, followed by post- Laramide drainage reversal that reused the older TM Figure 14. GoogleEarth image looking NW over the city of La Paz, Bolivia, onto the Alti- canyons (e.g., Potochnik, 2001; Scarborough, plano toward Lake Titicaca. Headward erosion of the La Paz River drainage has created 2001; Young, 2001b). The primary differences an asymmetrical drainage divide with steep SW slopes draining to the Atlantic Ocean and between these models for Grand Canyon and gentle NW slopes that lie within 200 m of the level of Lake Titicaca. the present interpretation are that (1) Grand Can- yon incision occurred prior to the main stage of Laramide deformation in the region, (2) drain- lakes Titicaca and Poopo to the west. These surface. By 16 Ma, at the onset of aggradation age reversal was Maastrichtian–Paleocene rather lakes are part of a closed drainage system of of the Bidahochi Formation NE of the divide, than mid-Tertiary, and (3) the early Tertiary interconnected lakes lying on a low-relief sur- the amphitheater lay at about the same elevation drainage was hydrologically connected to the face near 4000 m elevation, underlain by a thick as it was in early to mid-Tertiary time, but was Pacifi c Ocean rather than the interior. Tertiary basin fi ll that wedges out at roughly the instead rimmed by Permian strata and fl oored It should also be emphasized that placement position of the drainage divide (e.g., Zeilinger by Cambrian Tapeats Sandstone and underly- of the drainage divide near Lees Ferry in Paleo- and Schlunegger, 2007). From this confi gura- ing Proterozoic sedimentary and crystalline gene time, as opposed to a position farther SW tion, overtopping the divide would require very rocks, not greatly above its present erosion level in the eastern Mojave Desert, is not required by little adjustment of the landscape (the divide is (Figs. 13D and 13E; Flowers et al., 2008). Be- the coastal provenance data. Rather, the primary only 200 m higher than Lake Titicaca), but it tween 16 and 20 Ma, the Coconino Sandstone motivations for this facet of the model are that it would have the extraordinary consequence of di- would have been exposed to erosion, but would (1) terminates the high-relief, southern sediment verting a huge, long-lived upland drainage area have constituted only a small fraction (<10%) of supply (Mojave/Mogollon highlands) to the in- away from the closed lake basins and into the the total eroded volume from the eastern Grand terior basins in southern Utah, replacing it with Atlantic Ocean. As elaborated in the following, Canyon area, which underlay only a small frac- an asymmetric divide (Figs. 13B and 13C); a similar scenario may apply to the late Miocene tion (<10%) of the total area draining into the (2) places a high-relief, future spillover point integration that formed the Colorado River. Sespe basin. Given that ~30% of the zircons in near the geographic center of the plateau On the basis of this geomorphological the Coconino are younger than 1.3 Ga (Dickin- (Figs. 9B and 13F); and (3) maintains a consis- analogy , eastern Grand Canyon and its tribu- son and Gehrels, 2003), it is diffi cult to envisage tent fl ow direction within Grand Canyon at any taries would have formed amphitheater-like their contribution to Sespe zircon population as given time, rather than having to place a drain- headwaters cut into Mesozoic strata SW of the being more than ~0.3%. age divide within an already deep canyon, posi- divide (Fig. 9B), which from 55 to 20 Ma was Because thermochronologic evidence suggests tioned somewhere between Diamond Creek and rimmed by Upper Cretaceous strata near 2000 m that western Grand Canyon was likely cut to a the Upper Granite Gorge (Figs. 13B–13G). elevation and fl oored with Triassic near 500 m level near the Tapeats Sandstone by the Cali fornia elevation (Figs. 13C, 13D, and 13E). To avoid River, drainage reversal could well have supplied Tectonic Derangement of the Lower signifi cant input of 0.4–1.3 Ga zircons from abundant, highly survivable Tapeats gravels from Arizona River Drainage plateau erg deposits into the Sespe (Spafford Grand Canyon to the coast beginning in Paleo- Between 25 and 16 Ma, the Mogollon et al., 2009), I infer that the Mesozoic erg de- cene time, including the “lavender, pink, and Highlands, which lay between the Gila and posits were omitted across the sub-Cretaceous purple quartz arenite clasts” present in the basal upper Arizona rivers for most of Tertiary time unconformity in eastern Grand Canyon region, Simi Conglomerate of Paleocene age (Colburn (Fig. 9B), foundered due to crustal extension, such that mid-Cretaceous strata lay disconform- and Novak, 1989) and Eocene conglomerates of allowing much of the NE-directed drainage ably on the Triassic Chinle Formation. Prior to the Sespe Formation (Howard, 1996). I envisage previously fl owing toward Grand Canyon ca. 20 Ma, the Paleozoic erg deposits (Coconino a system wherein a main trunk stream gener- (Rim gravels) to be captured and to fl ow south- Sandstone) would have remained in the sub- ally transverse to the axis of Cor dilleran uplift westward (Mayer, 1979; Peirce et al., 1979;

24 Geological Society of America Bulletin, Month/Month 2010 Geological Society of America Bulletin, published online on 26 January 2011 as doi:10.1130/B30274.1

California River

Potochnik , 2001; Young, 2001b, 2008). This drainage with what is now the upper Colorado if not its precise dimensions, through kilometer- event resulted in ~1500 m of unroofi ng in a River drainage. The sudden increase in discharge scale of unroofi ng over tens of millions of 75-km-wide band along the NE fl ank of the Ari- and sediment load may have been as much as years (Flowers et al., 2008). To fi rst order, its zona homocline, including the eastern Grand two orders of magnitude, forcing a rapid cascade form appears to be independent of the amount Canyon region and portions of the future of spillover events that completed the integration of postincision unroofi ng, which is, at most, a Bidahochi basin region to the SE (Flowers et al., of the modern Colorado River drainage (e.g., few hundred meters in western Grand Canyon 2008). This event also corresponded to a transi- Meek and Douglass, 2001; Scarborough, 2001; but ~1500 m in eastern Grand Canyon. Its ba- tion from an aggradational regime on the plateau Spencer and Pearthree, 2001; House et al., 2008; sic form also appears to have survived a shift to one in which aggraded materials were largely Douglass et al., 2009). from predominantly humid, wet conditions at removed, with the informative exceptions of the One strength of this interpretation is that prior the time of incision to more arid conditions in Rim gravels (capped with early Miocene vol- to 6 Ma, the upper Colorado River drainage Oligocene and Miocene time (e.g., Gregory and canics) and Chuska erg deposits (e.g., Elston basin was either closed or drained out through Chase, 1994; Young, 1999). These aspects of the and Young, 1991; Cather et al., 2008). the northern plateau and the Pacifi c Northwest system contrast with Hack’s (1960) premise that By 17 Ma, extension in the Basin and Range (e.g., Spencer et al., 2008b), eliminating the tectonic and climatic forcing would, in general, Province had propagated northward to begin problem of fi nding a pre–6 Ma outlet (e.g., result in adjustment of the landscape to the new forming Grand Wash Trough. Owing to ex- Meek and Douglass, 2001). A second strength conditions. Any such adjustments are at present tension, by 13 Ma, successions in the coastal is that Grand Canyon was already in existence at beyond the resolution of the thermochronologi- delta system, eastern Mojave, and Grand Wash the time of spillover (e.g., Scarborough, 2001), cal data summarized in Figure 4. Trough had all developed pronounced angular eliminating the “precocious gully” problem of The history proposed here is not consistent unconformities (time between Figs. 13E and Hunt (1968, 1969). Any hypothetical river that with recent incision models based on extrapo- 13F; e.g., Davis, 1988; Dibblee, 1989; Wallace would overtop the Kaibab arch or take advan- lation of late Quaternary incision rates of 60– et al., 2005). Also by this time, rapid tectonism tage of a relatively shallow precursor canyon 190 m/m.y. back to 6 Ma, which are suffi cient to deranged the Arizona River and its tributaries cut through it is still confronted with another carve most of Grand Canyon (>1100 m in eastern below western Grand Canyon into a system 160 km of resistant, high plateau to the west Grand Canyon) since that time (e.g., Pederson of local basins, including the coastal region to carve through before arriving at the Grand et al., 2002b; Karlstrom et al., 2007, 2008). The (Fig. 14E; Ingersoll and Rumelhart, 1999). Wash Cliffs, with little hydrological impetus to total incision recorded by these measurements is This event left only Grand Canyon as the head- either incise Grand Canyon or begin excavation <100 m, covering a time period of 500–700 k.y. waters to Grand Wash Trough and Lake Huala- of the plateau interior. A high spillover point in If both the incision rates and the thermochrono- pai, and marked the end of the supply of exotic the Lees Ferry–Glen Canyon area would be ac- logical data are honored, this suggests that the Arizona River detritus to the coastal basins (e.g., companied by rapid knickpoint migration and relatively wet glacial climates over the last 2 m.y. Dibblee , 1989). This period also marked the on- kilometer-scale incision in order to establish the have had a signifi cant effect on the late Ceno- set of middle and upper Miocene aggradational present river grade, which is over 1000 m below zoic incision rate. For western Grand Canyon, events high on the Colorado Plateau as recorded the highest Bidahochi deposits to the east. The 130 m of incision could be accommodated at by the Bidahochi Formation, in Grand Wash profound lowering of base level at a position an average rate of 65 m/m.y. since 2 Ma, with Trough as recorded by the Muddy Creek For- well within the plateau interior (as opposed to rates averaging an order of magnitude less before mation, and along the coast as recorded by the along its margin at Grand Wash Cliffs) would that time. Similarly, erosion of 380 m in eastern Puente and equivalent formations (Fig. 14F). explain the thorough evacuation of any rela- Grand Canyon since 2 Ma is consistent with tively thin lacustrine deposits that may have ac- the AHE and AFT data. However, substantially Colorado River cumulated during Muddy Creek time in Grand more post–20 Ma erosion would require signifi - The interior of the plateau north of Grand Canyon (e.g., Meek and Douglass, 2001). In cantly different Cenozoic thermal histories for Canyon thus far has yielded much younger cool- summary, this interpretation relieves the long- the sample suites in the canyon and on the rim, ing ages than the SW margin of the plateau (e.g., standing twin headaches of searching for the contrary to the thermochronological data. Dur- Stöckli, 2005; Flowers et al., 2008), suggest- outlet of the pre–Grand Canyon Colorado ing Muddy Creek/Bidahochi time (13–6 Ma) or ing that major denudation, primarily of weakly River (e.g., Pederson, 2008)—no such entity Rim gravel/Chuska time (50–25 Ma), the can- resistant Mesozoic strata, has occurred since ever existed—and searching for a mechanism yon itself may have been aggrading (e.g., Scar- 10 Ma (Pederson et al., 2002a). After 6 Ma, to incise the Kaibab arch–Coconino terrace borough, 2001; Young, 2001b). Thus, even with under increasingly wet conditions in the Rocky region in late Cenozoic time (e.g., Karlstrom major unroofi ng near 20 Ma, modifi cation of its Mountains (e.g., Chapin, 2008), aggradation in et al., 2008)—Grand Canyon was already there. topographic form since 70 Ma appears to have the Bidahochi basin became suffi cient to overtop In so doing, it also suggests that the Colorado been relatively modest. the asymmetrical divide in the Lees Ferry–Glen River did not play a signifi cant role in excavat- From an historical perspective, referring to Canyon area (Fig. 13F), an event analogous to ing Grand Canyon. the Green, San Juan, and Colorado Rivers up- a hypothetical future overtopping of the asym- stream of Grand Wash Trough, Powell (1875, metric divide between the Altiplano and La CONCLUSIONS p. 198) rather forcefully concluded: Paz River drainages. This hypothesis is similar to that of Scarborough (2001), except that it Perhaps the primary implication of this in- Though the entire region has been folded and faulted places the spillover point somewhat farther north terpretive synthesis is that it reinforces the on a grand scale, these displacements have never de- than eastern Grand Canyon, to account for pre- counter intuitive conclusion that Grand Canyon termined the course of the streams. All the facts … lead to the inevitable conclusion that the system of Bidahochi cooling of the Upper Granite Gorge to is a long-lived equilibrium landscape of the drainage was determined antecedent to the faulting, near-surface temperatures (Flowers et al., 2008). general type envisaged by Hack (1960), having and folding, and erosion, which are observed, and This event integrated the upper Arizona River maintained its position and approximate depth, ante cedent, also, to the eruptive beds and cones.

Geological Society of America Bulletin, Month/Month 2010 25 Geological Society of America Bulletin, published online on 26 January 2011 as doi:10.1130/B30274.1

Wernicke

If the interpretation summarized in Figure 13 Billingsley, G.H., Beard, L.S., Priest, S.S., Wellmeyer, J.L., Davis, G.A., 1988, Rapid upward transport of mid-crustal is correct, then Grand Canyon was clearly cut and Block, D.L., 2004, Geologic Map of Lower Grand mylonitic gneisses in the footwall of a Miocene detach- Wash Cliffs and Vicinity, Mohave County, Northwest ment fault, Whipple Mountains, southeastern Califor- by an antecedent stream according to the defi ni- Arizona: U.S. Geological Survey Map MF-2427, scale nia: Geologische Rundschau, v. 77, p. 191–209, doi: tion of Powell (1875). However, if the Bidahochi 1:31,680, 1 sheet. 10.1007/BF01848684. Blackwelder, E., 1934, Origin of the Colorado River: Geo- Davis, S.J., Dickinson, W.R., Gehrels, G.E., Spencer, J.E., basin overtopped an asymmetrical drainage di- logical Society of America Bulletin, v. 45, p. 551–565. Lawton, T.F., and Carroll, A.R., 2010, The Paleogene vide, then prior to 6 Ma, any rivers upstream Blechschmidt, I., Matter, A., Preusser, F., and Rieke-Zapp, California River: Evidence of Mojave-Uinta paleo- from Grand Canyon were likely meandering D., 2009, Monsoon triggered formation of Qua- drainage from U-Pb ages of detrital zircons: Geology, ternary alluvial megafans in the interior of Oman: v. 38, p. 931–934, doi: 10.1130/G31250.1. on a high, low-relief plane (analogous to the Geomorphology, v. 110, p. 128–139, doi: 10.1016/j. Davis, W.M., 1899, The geographical cycle: The Geographi- meandering Desaguadero River in Bolivia that geomorph.2009.04.002. cal Journal, v. 14, p. 481–504, doi: 10.2307/1774538. connects Lake Titicaca and Lake Poopo), atop Bohannon, R.G., 1984, Nonmarine Sedimentary Rocks of DeCelles, P.G., 2004, Late Jurassic to Eocene evolution of Tertiary Age in the Lake Mead Region, Southeastern the Cordilleran thrust belt and foreland basin system, various post-Laramide deposits; therefore, they Nevada and Northwestern Arizona: U.S. Geological western USA: American Journal of Science, v. 304, would be considered superposed rather than Survey Professional Paper 1259, 72 p. p. 105–168, doi: 10.2475/ajs.304.2.105. Brady, R., Wernicke, B., and Fryxell, J., 2000, Kinematic Dibblee, T.W., 1989, Miocene conglomerates and breccias antecedent (e.g., Hunt, 1969). Invigoration of the evolution of a large-offset continental normal fault in the northeastern Ventura Basin and northern Los An- Colorado, Green, and San Juan Rivers by sud- system, South Virgin Mountain, Nevada: Geological geles Basin and their tectonic signifi cance, in Colburn, den base-level drop near Lees Ferry and their Society of America Bulletin, v. 112, p. 1375–1397, I.P., Abbott, P.L., and Minch, J., eds., Conglomerates doi: 10.1130/0016-7606(2000)112<1375:KEOALO> in Basin Analysis: A Symposium Dedidated to A.O. rapid adjustment to grade would have promoted 2.0.CO;2. Woodford: Bakersfi eld, California, Pacifi c Section, So- the development of the entrenched meanders of Burbank, D.W., Beck, R.A., Raynolds, R.G.H., Hobbs, R., ciety of Economic Paleontologists and Mineralogists, textbook fame that characterize all three rivers and Tahirkheli, R.A.K., 1988, Thrusting and gravel pro- p. 207–226. gradation in foreland basins—A test of post-thrusting Dickinson, W.R., 1981, Plate tectonics and the continental (e.g., Longwell, 1946; Scarborough, 2001). Al- gravel dispersal: Geology, v. 16, p. 1143–1146, margin of California, in Ernst, W.G., ed., The Geo- though Powell (1875) and other early workers doi: 10.1130/0091-7613(1988)016<1143:TAGPIF> tectonic Development of California, Rubey Volume I: 2.3.CO;2. Englewood Cliffs, New Jersey, Prentice-Hall, p. 1–28. did not envisage early Tertiary drainage reversal, Carslaw, H.S., and Jaeger, J.C., 1959, Conduction of Heat in Dickinson, W.R., 1983, Cretaceous sinistral strike slip along the present synthesis is, nonetheless, consonant Solids: Oxford, UK, Clarendon, 510 p. Nacimiento fault in coastal California: The American with Powell’s fundamental insight that a drain- Cather, S.M., 2004, Laramide orogeny in central and north- Association of Petroleum Geologists Bulletin, v. 67, ern New Mexico and southern Colorado, in Mack, p. 624–645. age divide across the Kaibab arch never existed. G.H., and Giles, K.A., eds., The Geology of New Dickinson, W.R., and Gehrels, G.E., 2003, U-Pb ages of Mexico, A Geologic History: New Mexico Geological detrital zircons from Permian and Jurassic eolian sand- ACKNOWLEDGMENTS Society Special Publication 11, p. 203–248. stones of the Colorado Plateau, USA: Paleogeographic Cather, S.M., Connell, S.D., Chamberlin, R.M., McIntosh, implications: Sedimentary Geology, v. 163, p. 29–66, The late Don Elston fi rst introduced me to the con- W.C., Jones, G.E., Potochnik, A.R., Lucas, S.G., and doi: 10.1016/S0037-0738(03)00158-1. cept of a Cretaceous age for Grand Canyon. I am also Johnson, P.S., 2008, The Chuska erg: Paleogeomorphic Dickinson, W.R., and Gehrels, G.E., 2008, Sediment deliv- grateful to J.M. Eiler, K.A. Farley, R.M. Flowers, K.W. and paleoclimatic implications of an Oligocene sand ery to the Cordilleran foreland basin: Insights from Huntington, J.B. Saleeby, and R.A. Young for discus- sea on the Colorado Plateau: Geological Society of U-Pb ages of detrital zircons in Upper Jurassic and sions that prompted this synthesis, and to S.J. Davis, America Bulletin, v. 120, p. 13–33. Cretaceous strata of the Colorado Plateau: American W.R. Dickinson, M. Grove, R.V. Ingersoll, and J.E. Chapin, C.E., 2008, Interplay of oceanographic and paleo- Journal of Science, v. 308, p. 1041–1082. climate events with tectonism during middle to late Dickinson, W.R., and Gehrels, G.E., 2009, U-Pb ages of Spencer for sharing detrital zircon data prior to pub- Miocene sedimentation across the southwestern detrital zircons in Jurassic eolian and associated sand- lication. I thank K.A. Farley and J. Harvey for assis- USA: Geosphere, v. 4, p. 976–991, doi: 10.1130/ stones of the Colorado Plateau: Evidence for trans- tance with the RDAAM model and using the HeFTy GES00171.1. continental dispersal and intraregional recycling of software. The presentation was greatly improved from Chipping, D.H., 1972, Early Tertiary paleogeography of sediment: Geological Society of America Bulletin, the careful and constructive reviews of S.M. Cather, central California: American Association of Petroleum v. 121, p. 408–433. R.V. Ingersoll, K.E. Karlstrom, P.K. Link, J. Pederson, Geologists Bulletin, v. 56, p. 480–493. Dickinson, W.R., Klute, M.A., Hayes, M.J., Janecke, S.U., J.D. Walker, and R.A. Young, although responsibility Chisholm, T.J., and Chapman, D.S., 1992, Climate change Lundin, E.R., Mckittrick, M.A., and Olivares, M.D., for errors in either fact or interpretation rest solely inferred from analysis of borehole temperatures—An 1988, Paleogeographic and paleotectonic setting of example from western Utah: Journal of Geophysical Laramide sedimentary basins in the central Rocky with the author. Figure 13 was drafted by J. Mayne. Research–Solid Earth, v. 97, p. 14,155–14,175, doi: Mountain region: Geological Society of America Bul- This research was funded by National Science Foun- 10.1029/92JB00765. letin, v. 100, p. 1023–1039, doi: 10.1130/0016-7606 dation grant EAR-0810324 and the Gordon and Betty Colburn, I.P., and Novak, G.A., 1989, Paleocene conglom- (1988)100<1023:PAPSOL>2.3.CO;2. Moore Foundation (Tectonics Obervatory Cont. #143). erates of the Santa Monica Mountains, California: Dickinson, W.R., Ducea, M., Rosenberg, L.I., Greene, Petrology, stratigraphy, and environments of depo- H.G., Graham, S.A., Clark, J.C., Weber, G.E., Kidder, REFERENCES CITED sition, in Colburn, I.P., Abbott, P.L., and Minch, J., S., Ernst, W.G., and Brabb, E.E., 2005, Net Dextral eds., Conglomerates in Basin Analysis: A Symposium Slip, Neogene San Gregorio–Hosgri Fault Zone, Atwater, T., and Stock, J., 1998, Pacifi c–North America Dedicated to A.O. Woodford: Bakersfi eld, California, Coastal California: Geologic Evidence and Tectonic plate tectonics of the Neogene southwestern United Pacifi c Section, Society of Economic Paleontologists Implications: Geological Society of America Special States: An update: International Geology Review, and Mineralogists, p. 227–253. Paper 391, 43 p. v. 40, p. 375–402, doi: 10.1080/00206819809465216. Critelli, S., Rumelhart, P.E., and Ingersoll, R.V., 1995, Petro- Dorsey, R.J., Fluette, A., McDougall, K., Housen, B.A., Beard, L.S., 1993, Tertiary stratigraphy of the South Virgin facies and provenance of the Puente Formation (Middle Janecke, S.U., Axen, G.J., and Shirvell, C.R., 2007, Mountains, southeast Nevada and the Grand Wash to Upper Miocene), Los Angeles Basin, Southern Cali- Chronology of Miocene-Pliocene deposits at Split Trough, northwest Arizona, in Sherrod, D.R., and fornia—Implications for rapid uplift and accumulation Mountain Gorge, Southern California: A record of re- Nielson, J.E., eds., Tertiary Stratigraphy of Highly Ex- rates: Journal of Sedimentary Research, ser. A, Sedi- gional tectonics and Colorado River evolution: Geol- tended Terranes, California, Arizona and Nevada: U.S. mentary Petrology and Processes, v. 65, p. 656–667. ogy, v. 35, p. 57–60, doi: 10.1130/G23139A.1. Geological Survey Bulletin 2053, p. 29–32. Crow, R., Karlstrom, K.E., McIntosh, W., Peters, L., and Douglass, J., Meek, N., Dorn, R.I., and Schmeeckle, M.W., Bernet, M., 2009, A fi eld-based estimate of the zircon fi ssion- Dunbar, N., 2008, History of Quaternary volcanism 2009, A criteria-based methodology for determining track closure temperature: Chemical Geology, v. 259, and lava dams in western Grand Canyon based on the mechanism of transverse drainage development, p. 181–189, doi: 10.1016/j.chemgeo.2008.10.043. Lidar analysis, Ar-40/Ar-39 dating, and fi eld studies: with application to the southwestern United States: Billingsley, G.H., Hawkins, D., Huntoon, P.W., Ilg., B.R., Implications for fl ow stratigraphy, timing of volcanic Geological Society of America Bulletin, v. 121, Sears, J.H., and Williams, M.L.,1996, Geologic Map events, and lava dams: Geosphere, v. 4, p. 183–206, p. 586–598. of the Eastern Part of the Grand Canyon National Park: doi: 10.1130/GES00133.1. Dumitru, T.A., Duddy, I.R., and Green, P.F., 1994, Meso- Grand Canyon, Arizona, Grand Canyon, Grand Can- Dallegge, T.A., Ort, M.H., McIntosh, W.C., and Perkins, zoic-Cenozoic burial, uplift, and erosion history of the yon Association, scale 1:65,500, 1 sheet. M.E., 2001, Age and depositional basin morphology west-central Colorado Plateau: Geology, v. 22, p. 499– Billingsley, G.H., Wenrich, K.J., Huntoon, P.W., and Young, of the Bidahochi Formation and implications for the 502, doi: 10.1130/0091-7613(1994)022<0499:MCBU R.A., 1999, Breccia-Pipe and Geologic Map of the ancestral upper Colorado River, in Young, R.A., and AE>2.3.CO;2. Southwestern Part of the Hualapai Indian Reservation Spamer, E.E., eds., Colorado River Origin and Evo- Dutton, C.E., 1882, Tertiary History of the Grand Cañon and Vicinity, Arizona: U.S. Geological Survey Map lution: Grand Canyon, Arizona, Grand Canyon Asso- District: Reston, Virginia, U.S. Geological Survey, I-2554, scale 1:48,000, 1 sheet. ciation, p. 47–51. 92 p.

26 Geological Society of America Bulletin, Month/Month 2010 Geological Society of America Bulletin, published online on 26 January 2011 as doi:10.1130/B30274.1

California River

Eiler, J.M., 2007, “Clumped-isotope” geochemistry—The logi cal Society of America Bulletin, v. 106, p. 145–154, River Origin and Evolution: Grand Canyon, Arizona, study of naturally-occurring, multiply-substituted doi: 10.1130/0016-7606(1994)106<0145:TDOUCT> Grand Canyon Association, p. 101–106. isotopologues: Earth and Planetary Science Letters, 2.3.CO;2. Howard, K.A., Lundstrom, S.C., Malmon, D.V., and Hook, v. 262, p. 309–327, doi: 10.1016/j.epsl.2007.08.020. Green, P.F., 1988, The relationship between track shorten- S.J., 2008, Age distribution and formation of late Elston, D.P., and Young, R.A., 1991, Cretaceous-Eocene ing and fi ssion-track age reduction in apatite—Com- Ceno zoic paleovalleys of the lower Colorado River and (Laramide) landscape development and Oligocene- bined infl uences of inherent instability, annealing their relation to river aggradation and degradation, in Pliocene drainage reorganization of transition zone anisotropy, length bias and system calibration: Earth Reheis, M.C., Hershler, R., and Miller, D.M., eds., Late and Colorado Plateau, Arizona: Journal of Geophysi- and Planetary Science Letters, v. 89, p. 335–352, doi: Cenozoic Drainage History of the Southwestern Great cal Research–Solid Earth and Planets, v. 96, p. 12,389– 10.1016/0012-821X(88)90121-5. Basin and Lower Colorado River Basin: Geologic and 12,406, doi: 10.1029/90JB01978. Gregory, K.M., and Chase, C.G., 1994, Tectonic and climatic Biotic Perspectives: Geological Society of America Elston, D.P., Young, R.A., McKee, E.H., and Dennis, M.L., signifi cance of a late Eocene low-relief, high-level Special Paper 439, p. 391–410. 1989, Paleontology, clast ages and paleomagnetism of geomorphic surface, Colorado: Journal of Geophysi- Hunt, C.B., 1968, Review of “Evolution of the Colorado Upper Paleocene and Eocene gravel and limestone de- cal Research–Solid Earth, v. 99, p. 20,141–20,160, doi: River in Arizona,” in McKee, E.D., Wilson, R.F., posits, Colorado Plateau and transition zone, northern 10.1029/94JB00132. Breed, W.J., and Breed, C.S., eds.: Geotimes, v. 13, no. 2, and central Arizona, in Elston, D.P., Billingsley, G.H., Grove, K., 1989, Upper Cretaceous conglomerates from the p. 39. and Young, R.A., eds., Geology of Grand Canyon, Salinian terrane, west-central California, in Colburn, Hunt, C.B., 1969, Geologic history of the Colorado River: Northern Arizona (with Colorado River Guides): Wash- I.P., Abbott, P.L., and Minch, J., eds., Conglomerates The Colorado River region and John Wesley Powell: ington, D.C., American Geophysical Union, p. 155–165. in Basin Analysis: A Symposium Dedidated to A.O. U.S. Geological Survey Professional Paper 669C, Emmons, S.F., 1897, The origin of the Green River: Science, Woodford: Bakersfi eld, California, Pacifi c Section, So- p. 59–130. v. 6, p. 19–21, doi: 10.1126/science.6.131.19. ciety of Economic Paleontologists and Mineralogists, Huntington, K.W., Wernicke, B.P., and Eiler, J.M., 2010, Farley, K.A., 2000, Helium diffusion from apatite: General p. 143–160. The infl uence of climate change and uplift on Colo- behavior as illustrated by Durango fl ourapatite: Journal Grove, K., 1993, Latest Cretaceous basin formation within rado Plateau paleotemperatures from carbonate of Geophysical Research, v. 105, p. 2903–2914, doi: the Salinian terrane of west-central California: Geo- ‘clumped isotope’ thermometry: Tectonics, v. 29, doi: 10.1029/1999JB900348. logi cal Society of America Bulletin, v. 105, p. 447–463, 10.1029/2009TC002449. Faulds, J.E., Wallace, M.A., Gonzales, L.A., and Heizler, doi: 10.1130/0016-7606(1993)105<0447:LCBFWT> Ingersoll, R.V., 1982, Initiation and evolution of the Great M.T., 2001, Depositional environment and paleogeo- 2.3.CO;2. Valley forearc basin of northern and central Califor- graphic implications of the late Miocene Hualapai Hack, J.T., 1960, Interpretation of erosional topography in nia, in Leggett, J.K., ed., Trench-Forearc Geology: Limestone, northwestern Arizona, in Young, R.A., and humid temperate regions: American Journal of Sci- Sedimentation and Tectonics on Modern and Ancient Spamer, E.E., eds., Colorado River Origin and Evolu- ence, v. 258–A, p. 80–97. Active Plate Margins: Geological Society of London tion: Grand Canyon, Arizona, Grand Canyon Asso- Heimsath, A.M., Dietrich, W.E., Nishiizumi, K., and Finkel, Special Publication 10, p. 459–467. ciation, p. 81–88. R.C., 1999, Cosmogenic nuclides, topography, and the Ingersoll, R.V., 1998, Phanerozoic tectonic evolution of Finnell, T.L., 1966, Geologic Map of the Cibeque Quadran- spatial variation of soil depth: Geomorphology, v. 27, central California and environs, in Ernst, W.G., and gle, Navajo County, Arizona: U.S. Geological Survey p. 151–172, doi: 10.1016/S0169-555X(98)00095-6. Nelson , C.A., eds., Integrated Earth and Environmental Map GQ-545, Geological Quadrangle Series, scale Henry, C.D., 2008, Ash-fl ow tuffs and paleovalleys in north- Evolution of the Southwestern United States: Colum- 1:62,500, 1 sheet. eastern Nevada: Implications for Eocene paleogeo- bia, Maryland, Bellwether, p. 349–364. Fitzgerald, P.G., Fryxell, J.E. and Wernicke, B.P., 1991, graphy and extension in the Sevier hinterland, northern Ingersoll, R.V., and Rumelhart, P.E., 1999, Three-stage evo- Miocene crustal extension and uplift in southeastern Great Basin: Geosphere, v. 4, p. 1–35, doi: 10.1130/ lution of the Los Angeles basin, southern California: Nevada—Constraints from fission-track analysis: GES00122.1. Geology, v. 27, p. 593–596, doi: 10.1130/0091-7613 Geol ogy, v. 19, p. 1013–1016. Hill, C.A., and Ranney, W.D., 2008, A proposed Laramide (1999)027<0593:TSEOTL>2.3.CO;2. Fitzgerald, P.G., Duebendorfer, E.M., Faulds, J.E., and proto–Grand Canyon: Geomorphology, v. 102, p. 482– Jacobson, C.E., Grove, M., Pedrick, J.N., Barth, A.P., O’Sullivan, P., 2009, South Virgin–White Hills detach- 495, doi: 10.1016/j.geomorph.2008.05.039. Marsaglia, K.M., Gehrels, G.E., and Nourse, J.A., ment fault system of SE Nevada and NW Arizona: Hill, C.A., Eberz, N., and Buecher, R.H., 2008, A karst 2010, Late Cretaceous–early Cenozoic tectonic evolu- Applying apatite fi ssion track thermochronology to connection model for Grand Canyon, Arizona, USA: tion of the southern California margin inferred from constrain the tectonic evolution of a major continen- Geomorphology, v. 95, p. 316–334, doi: 10.1016/ provenance of trench and forearc sediments: Geologi- tal detachment fault: Tectonics, v. 28, p. TC2001, doi: j.geomorph.2007.06.009. cal Society of America Bulletin, doi: 10.1130/B30238. 10.1029/2007TC002194. Holm, R.F., 2001, Cenozoic paleogeography of the central Jinnah, Z.A., Roberts, E.M., Deino, A.L., Larsen, J.S., Link, Flowers, R.M., Shuster, D.L., Wernicke, B.P., and Farley , Mogollon Rim–southern Colorado Plateau region, Ari- P.K., and Fanning, C.M., 2009, New Ar-40-Ar-39 K.A., 2007, Radiation damage control on apatite zona, revealed by Tertiary gravel deposits, Oligocene and detrital zircon U-Pb ages for the Upper Creta- (U-Th)/He dates from the Grand Canyon region, Colo- to Pleistocene lava fl ows, and incised streams: Geologi- ceous Wahweap and Kaiparowits Formations on the rado Plateau: Geology, v. 35, p. 447–450, doi: 10.1130/ cal Society of America Bulletin, v. 113, p. 1467–1485, Kaiparowits Plateau, Utah: Implications for regional G23471A.1. doi: 10.1130/0016-7606(2001)113<1467:CPOTCM> correlation, provenance, and biostratigraphy: Cre- Flowers, R.M., Wernicke, B.P., and Farley, K.A., 2008, Un- 2.0.CO;2. taceous Research, v. 30, p. 287–299, doi: 10.1016/ roofi ng, incision, and uplift history of the southwestern House, M.A., Wernicke, B.P., and Farley, K.A., 1998, Dat- j.cretres.2008.07.012. Colorado Plateau from apatite (U-Th)/He thermo- ing topography of the Sierra Nevada, California, using Karlstrom, K.E., Crow, R.S., Peters, L., McIntosh, W., chronometry: Geological Society of America Bulletin, apatite (U-Th)/He ages: Nature, v. 396, p. 66–69, doi: Raucci, J., Crossey, L.J., Umhoefer, P., and Dunbar, v. 120, p. 571–587, doi: 10.1130/B26231.1. 10.1038/23926. N., 2007, Ar-40/Ar-39 and fi eld studies of Quaternary Flowers, R.M., Ketcham, R.A., Shuster, D.L., and Farley, House, M.A., Wernicke, B.P., and Farley, K.A., 2001, basalts in Grand Canyon and model for carving Grand K.A., 2009, Apatite (U-Th)/He thermochronometry Paleo-geomorphology of the Sierra Nevada, California, Canyon: Quantifying the interaction of river incision using a radiation damage accumulation and anneal- from (U-Th)/He ages in apatite: American Journal of and normal faulting across the western edge of the ing model: Geochimica et Cosmochimica Acta, v. 73, Science, v. 301, p. 77–102, doi: 10.2475/ajs.301.2.77. Colorado Plateau: Geological Society of America Bul- p. 2347–2365, doi: 10.1016/j.gca.2009.01.015. House, P.K., Pearthree, P.A., and Perkins, M.E., 2008, Strati- letin, v. 119, p. 1283–1312, doi: 10.1130/0016-7606 Foster, D.A., Gleadow, A.J.W., Reynolds, S.J., and Fitz- graphic evidence for the role of lake spillover in the in- (2007)119[1283:AAFSOQ]2.0.CO;2. gerald, P.G., 1993, Denudation of metamorphic core ception of the lower Colorado River in southern Nevada Karlstrom, K.E., Crow, R., Crossey, L.J., Coblentz, D., and complexes and the reconstruction of the transition and western Arizona, in Reheis, M.C., Hershler, R., and Van Wijk, J.W., 2008, Model for tectonically driven in- zone, west central Arizona—Constraints from apatite Miller, D.M., eds., Late Cenozoic Drainage History cision of the younger than 6 Ma Grand Canyon: Geol- fi ssion-track thermochronology: Journal of Geophysi- of the Southwestern Great Basin and Lower Colorado ogy, v. 36, p. 835–838, doi: 10.1130/G25032A.1. cal Research–Solid Earth, v. 98, p. 2167–2185, doi: River Basin: Geologic and Biotic Perspectives: Geo logi- Kelley, S.A., and Chapin, C.E., 2004, Denudation history 10.1029/92JB02407. cal Society of America Special Paper 439, p. 335–353. and internal structure of the Front Range and Wet Ghosh, P., Garzione, C.N., and Eiler, J.M., 2006, Rapid up- Howard, J.L., 1996, Paleocene to Holocene paleodeltas Mountains, Colorado, based on apatite-fi ssion-track lift of the Altiplano revealed through C-13–O-18 bonds of ancestral Colorado River offset by the San An- thermochronology, in Cather, S.M., McIntosh, W.C., in paleosol carbonates: Science, v. 311, p. 511–515, dreas fault system, southern California: Geology, and Kelley, S.A., eds., Tectonics, Geochronology and doi: 10.1126/science.1119365. v. 24, p. 783–786, doi: 10.1130/0091-7613(1996)024 Volcanism in the Southern Rocky Mountains and Rio Goldstrand, P.M., 1990, Stratigraphy and Paleogeography <0783:PTHPOA>2.3.CO;2. Grande Rift: Socorro, New Mexico, New Mexico Bu- of Late Cretaceous and Paleogene Rocks of Southwest Howard, J.L., 2000, Provenance of quartzite clasts in reau of Geology and Mineral Resources Bulletin 160, Utah: Utah Geological and Mineralogical Survey Mis- the Eocene-Oligocene Sespe Formation: Paleogeo- p. 41–78. cellaneous Publication 90–2, 58 p. graphic implications for southern California and Kelley, S.A., Chapin, C.E., and Karlstrom, K.E., 2001, Goldstrand, P.M., 1992, Evolution of Late Cretaceous and the ancestral Colorado River: Geological Society of Laramide cooling histories of Grand Canyon, Arizona, Early Tertiary basins of southwest Utah based on clas- America Bulletin, v. 112, p. 1635–1649, doi: 10.1130/ and the Front Range, Colorado, determined from apa- tic petrology: Journal of Sedimentary Petrology, v. 62, 0016-7606(2000)112<1635:POQCIT>2.0.CO;2. tite fi ssion-track thermochronology, in Young, R.A., p. 495–507. Howard, K.A., and Bohannon, R.G., 2001, Lower Colorado and Spamer, E.E., eds., Colorado River Origin and Goldstrand, P.M., 1994, Tectonic development of Upper Cre- River: Upper Cenozoic deposits, incision and evolu- Evolution: Grand Canyon, Arizona, Grand Canyon taceous to Eocene strata of southwestern Utah: Geo- tion, in Young, R.A., and Spamer, E.E., eds., Colorado Asso ciation, p. 37–42.

Geological Society of America Bulletin, Month/Month 2010 27 Geological Society of America Bulletin, published online on 26 January 2011 as doi:10.1130/B30274.1

Wernicke

Ketcham, R.A., Donelick, R.A., and Carlson, W.D., 1999, Martin, M.W., and Walker, J.D., 1992, Extending the west- Colorado River Origin and Evolution: Grand Canyon, Variability of apatite fi ssion-track annealing kinetics: ern North American Proterozoic and Paleozoic craton Arizona, Grand Canyon Association, p. 17–22. III. Extrapolation to geological time scales: The Amer- through the Mojave Desert: Geology, v. 20, p. 753–756, Potochnik, A.R., and Faulds, J.E., 1998, A tale of two rivers: ican Mineralogist, v. 84, p. 1235–1255. doi: 10.1130/0091-7613(1992)020<0753:ETWNAP> Mid-Tertiary structural inversion and drainage reversal Kies, R.P., and Abbott, P.L., 1983, Rhyolite clast populations 2.3.CO;2. across the southern boundary of the Colorado Plateau, and tectonics in the California continental borderland: Mayer, L., 1979, Evolution of the Mogollon Rim in cen- in Duebendorfer, E.M., ed., Geologic Excursions in Journal of Sedimentary Petrology, v. 53, p. 461–476. tral Arizona: Tectonophysics, v. 61, p. 49–62, doi: Northern and Central Arizona, Field Trip Guidebook: Kimbrough, D., Grove, M., Gehrels, G., Dorsey, R., House, 10.1016/0040-1951(79)90291-9. Flagstaff, Arizona, Geological Society of America, K.P., Howard, K., Pearthree, P.A., Spencer, J.E., and McKee, E.D., Wilson, R.F., Breed, W.J., and Breed, C.S., Rocky Mountain Section, p. 149–174. Mahoney, B., 2007, Detrital zircon record of Colorado 1967, Evolution of the Colorado River in Arizona: Powell, J.L., 2005, Grand Canyon, Solving Earth’s Grand- River incision: Eos (Transactions, American Geophysi- Flagstaff, Arizona, Museum of Northern Arizona Bul- est Puzzle: New York, Pearson Education, Inc., 308 p. cal Union), v. 88, no. T41, p. C–03. letin 44, 67 p. Powell, J.W., 1875, Exploration of the Colorado River of the Lachenbruch, A.H., and Sass, J.H., 1978, Models of an McQuarrie, N., and Wernicke, B.P., 2005, An animated West and Its Tributaries: Washington, D.C., U.S. Gov- extending lithosphere and heat fl ow in the Basin and tectonic reconstruction of southwestern North Amer- ernment Printing Offi ce, 291 p. Range Province, in Smith, R.B., and Eaton, G.P., eds., ica since 36 Ma: Geosphere, v. 1, p. 147–172, doi: Quigley, M., Karlstrom, K., Kelley, S., and Heizler, M., Cenozoic Tectonics and Regional Geophysics of the 10.1130/GES00016.1. 2010, Timing and mechanisms of basement uplift Western Cordillera: Geological Society of America Meek, N., and Douglass, J., 2001, Lake overfl ow: An alter- and exhumation in the Colorado Plateau–Basin and Memoir 152, p. 209–250. native hypothesis for Grand Canyon incision and de- Range Transition Zone, Virgin Mountain anticline, Larsen, J.S., 2007, Facies and Provenance of the Pine Hol- velopment of the Colorado River, in Young, R.A., and Nevada-Arizona, in Umhoefer, P.J., Beard, L.S. and low Formation: Implications for the Sevier Foreland Spamer, E.E., eds., Colorado River Origin and Evolu- Lamb, M.A., eds., Miocene Tectonics of the Lake Basin Evolution and the Paleocene Climate of South- tion: Grand Canyon, Arizona, Grand Canyon Associa- Mead Region, Central Basin and Range: Geological ern Utah [M.S. thesis]: Pocatello, Idaho State Univer- tion, p. 199–204. Society of America Special Paper 463, p. 311–329, doi: sity, 119 p. Naeser, C.W., Duddy, I.R., Elston, D.P., Dumitru, T.A., and 10.1130/2010.2463(14). Larsen, J.S., Link, P.K., Roberts, E.M., Tapanila, L., and Green, P.F., 1989, Fission-track dating: Ages for Cam- Reiners, P.W., and Shuster, D.L., 2009, Thermochronology Fanning, C.M., 2010, Cyclic stratigraphy of the Paleo- brian strata, and Laramide and post–middle Eocene and landscape evolution: Physics Today, v. 62, p. 31– gene Pine Hollow Formation and detrital zircon prov- cooling events from the Grand Canyon, Arizona, in 36, doi: 10.1063/1.3226750. enance of Campanian to Eocene sandstones of the Elston, D.P., Billingsley, G.H., and Young, R.A., eds., Reiners, P.W., Brady, R., Farley, K.A., Fryxell, J.E., Wer- northwestern Kaiparowits Basin, south-central Utah: Geology of Grand Canyon, Northern Arizona (with nicke, B.P., and Lux, D., 2000, Helium and argon Utah Geological Association Field Trip Guidebook (in Colorado River Guides): Washington, D.C., American thermochronometry of the Gold Butte block, South press). Geophysical Union, p. 139–144. Virgin Mountains, Nevada: Earth and Planetary Laslett, G.M., Green, P.F., Duddy, I.R., and Gleadow, Niemi, N.A., Wernicke, B.P., Brady, R.J., Saleeby, J.B., Science Letters, v. 178, p. 315–326, doi: 10.1016/ A.J.W., 1986, Thermal annealing of fission tracks in and Dunne, G.C., 2001, Distribution and provenance S0012-821X(00)00080-7. apatite: 2. A quantitative analysis: Chemical Geology, of the middle Miocene Eagle Mountain Formation, Renik, B., Christie-Blick, N., Troxel, B.W., Wright, L.A., v. 65, p. 1–13. and implications for regional kinematic analysis and Niemi, N.A., 2008, Re-evaluation of the middle Lawton, T.F., Pollock, S.L., and Robinson, R.A.J., 2003, of the Basin and Range Province: Geological So- Miocene Eagle Mountain Formation and its sig- Integrating sandstone petrology and nonmarine se- ciety of America Bulletin, v. 113, p. 419–442, doi: nifi cance as a piercing point for the interpretation of quence stratigraphy: Application to the Late Creta- 10.1130/0016-7606(2001)113<0419:DAPOTM> extreme extension across the Death Valley region, Cali- ceous fl uvial systems of southwestern Utah: Journal 2.0.CO;2. fornia, USA: Journal of Sedimentary Research, v. 78, of Sedimentary Research, v. 73, p. 389–406, doi: Page, B.M., 1981, The southern Coast Ranges, in Ernst, p. 199–219, doi: 10.2110/jsr.2008.022. 10.1306/100702730389. W.G., ed., The Geotectonic Development of Califor- Reynolds, M.W., 1969, Stratigraphy and Structural Geology Link, P.K., Roberts, E., Fanning, C.M., and Larsen, J.S., nia, Rubey Volume I: Englewood Cliffs, New Jersey, of the Titus and Titanothere Canyons Area, Death Val- 2007, Detrital zircon age populations from Upper Cre- Prentice-Hall, p. 329–417. ley, California [Ph.D. thesis]: Berkeley, University of taceous and Paleogene Wahweap, Kaiparowits, Canaan Pearthree, P.A., Spencer, J.E., Faulds, J.E., and House, P.K., California, 310 p. Peak, Pine Hollow and Claron Formations, Kaiparo- 2008, Comment on “Age and evolution of the Grand Sahagian, D., Proussevitch, A., and Carlson, W., 2003, wits Plateau, southern Utah: Geological Society of Canyon revealed by U-Pb dating of water table–type Timing of Colorado Plateau uplift: Initial constraints America Abstracts with Programs, v. 39, no. 5, p. 7. speleothems”: Science, v. 321, p. 1634c, doi: 10.1126/ from vesicular basalt-derived paleoelevations: Geol- Longwell, C.R., 1936, Geology of the Boulder Reservoir science.1158862. ogy, v. 31, p. 807–810, doi: 10.1130/0091-7613 fl oor, Arizona-Nevada: Geological Society of America Pederson, J., 2008, The mystery of the pre–Grand Can- (2003)031<0192:>2.0.CO;2. Bulletin, v. 47, p. 1393–1476. yon Colorado River—Results from the Muddy Creek Saleeby, J., 2003, Segmentation of the Laramide slab—Evi- Longwell, C.R., 1946, How old is the Colorado River: Formation: GSA Today, v. 18, p. 4–10, doi: 10.1130/ dence from the southern Sierra Nevada region: Geolog- American Journal of Science, v. 244, p. 817–835. GSAT01803A.1. ical Society of America Bulletin, v. 115, p. 655–668, Lowry, A.R., and Smith, R.B., 1995, Strength and rheology Pederson, J., Mackley, R.D., and Eddleman, J.L., 2002a, doi: 10.1130/0016-7606(2003)115<0655:SOTLSF> of the western US Cordillera: Journal of Geophysical Colorado Plateau uplift and erosion evaluated us- 2.0.CO;2. Research-Solid Earth, v. 100, p. 17,949–17,963. ing GIS: GSA Today, v. 12, p. 4–10, doi: 10.1130/ Sass, J.H., Lachenbruch, A.H., Galanis, S.P., Morgan, P., Lucchitta, I., 1966, Cenozoic Geology of the Lake Mead 1052-5173(2002)012<0004:CPUAEE>2.0.CO;2. Priest, S.S., Moses, T.H., and Munroe, R.J., 1994, Area Adjacent to the Grand Wash Cliffs, Arizona Pederson, J., Karlstrom, K.E., McIntosh, W.C., and Sharp, Thermal regime of the southern Basin and Range Prov- [Ph.D. thesis]: State College, Pennsylvania State Uni- W., 2002b, Geomorphic implications of new absolute ince: 1. Heat-fl ow data from Arizona and the Mojave versity, 218 p. ages and associated incision and fault-slip rates in Grand Desert of California and Nevada: Journal of Geophysi- Lucchitta, I., 1972, Early history of Colorado River and Canyon: Geology, v. 30, p. 739–742, doi: 10.1130/ cal Research–Solid Earth, v. 99, p. 22,093–22,119, doi: Basin and Range Province: Geological Society of 0091-7613(2002)030<0739:DIOTGC>2.0.CO;2. 10.1029/94JB01891. America Bulletin, v. 83, p. 1933–1948, doi: 10.1130/ Pederson, J., Young, R., Lucchitta, I., Beard, L.S., and Bill- Scarborough, R., 2001, Neogene development of the Little 0016-7606(1972)83[1933:EHOTCR]2.0.CO;2. ingsley, G., 2008, Comment on “Age and evolution of Colorado River Valley and eastern Grand Canyon: Lucchitta, I., 1979, Late Cenozoic uplift of the southwest- the Grand Canyon revealed by U-Pb dating of water Field evidence for an overtopping hypothesis, in ern Colorado Plateau and adjacent lower Colorado table–type speleothems”: Science, v. 321, p. 1634b, Young, R.A., and Spamer, E.E., eds., Colorado River River region: Tectonophysics, v. 61, p. 63–95, doi: doi: 10.1126/science.1158019. Origin and Evolution: Grand Canyon, Arizona, Grand 10.1016/0040-1951(79)90292-0. Peirce, H.W., Damon, P.E., and Shafi qullah, M., 1979, An Canyon Association, p. 207–212. Lucchitta, I., 1984, Development of landscape in northwest Oligocene (?) Colorado Plateau edge in Arizona: Tec- Shuster, D.L., and Farley, K.A., 2004, He-4/He-3 thermo- Arizona: The country of plateaus and canyons, in tonophysics, v. 61, p. 1–24, doi: 10.1016/0040-1951 chronometry: Earth and Planetary Science Letters, Smiley, T.L., Nations, J.D., Péwé, T.L., and Schafer, (79)90289-0. v. 217, p. 1–17, doi: 10.1016/S0012-821X(03)00595-8. J.P., eds., Landscapes of Arizona, the Geologic Story: Polyak, V., Hill, C., and Asmerom, Y., 2008a, Age and evo- Shuster, D.L., Flowers, R.M., and Farley, K.A., 2006, The Lanham, Maryland, University Press of America, lution of the Grand Canyon revealed by U-Pb dating infl uence of natural radiation damage on helium dif- p. 269–302. of water table–type speleothems: Science, v. 319, fusion kinetics in apatite: Earth and Planetary Sci- Lucchitta, I., 2003, History of the Grand Canyon and the p. 1377–1380, doi: 10.1126/science.1151248. ence Letters, v. 249, p. 148–161, doi: 10.1016/ Colorado River, in Beus, S.S., and Morales, M., eds., Polyak, V., Hill, C., and Asmerom, Y., 2008b, Response to j.epsl.2006.07.028. Grand Canyon Geology: New York, Oxford University comments on the “Age and evolution of the Grand Snow, J.K., and Lux, D.R., 1999, Tectono-sequence stratig- Press, p. 260–274. Canyon revealed by U-Pb dating of water table–type raphy of Tertiary rocks in the Cottonwood Mountains Lucchitta, I., and Jeanne, R.A., 2001, Geomorphic features speleothems”: Science, v. 321, p. 1634d, doi: 10.1126/ and northern Death Valley area, California and Nevada, and processes of the Shivwits Plateau, Arizona, and science.1159762. in Wright, L.A., and Troxel, B.W., eds., Cenozoic Ba- their constraints on the age of western Grand Canyon, Potochnik, A.R., 2001, Paleogeomorphic evolution of sins of the Death Valley Region: Geological Society of in Young, R.A., and Spamer, E.E., eds., Colorado River the Salt River region: Implications for Cretaceous- America Special Paper 333, p. 17–64. Origin and Evolution: Grand Canyon, Arizona, Grand Laramide inheritance for ancestral Colorado River Spafford, C.D., Ingersoll, R.V., and Grove, M., 2009, Canyon Association, p. 65–69. drainage, in Young, R.A., and Spamer, E.E., eds., Detrital zircons and sandstone petrofacies of the Sespe

28 Geological Society of America Bulletin, Month/Month 2010 Geological Society of America Bulletin, published online on 26 January 2011 as doi:10.1130/B30274.1

California River

Formation, southern California: Ventura, California, supply limitation and the infl uence of Glen Canyon model to distinguish between the causes of rapid sym- Pacifi c Section Meeting, May 3–5, 2009, American Dam: Water Resources Research, v. 36, p. 515–542, metrical scarp retreat and scarp dissection, in Hack, Association of Petroleum Geologists, Pacifi c Section doi: 10.1029/1999WR900285. J.T., and Morisawa, M., eds., Tectonic Geomorphol- Meeting, Search and Discovery article 90088. Turcotte, D.L., and Schubert, G., 1982, Geodynamics: New ogy: London, Allen and Unwin, p. 261–278. Spencer, J.E., and Pearthree, P.A., 2001, Headward erosion York, John Wiley and Sons, 450 p. Young, R.A., 1999, Appendix: Nomenclature and ages of versus closed-basin spillover as alternative causes of Walcott, C.D., 1890, Study of a line of displacement in the Late Cretaceous(?)–Tertiary strata in the Hualapai Pla- Neogene capture of the ancestral Colorado River by the Grand Canyon of the Colorado: Geological Society of teau region, northwest Arizona, in Billingsley, G.H., Gulf of California, in Young, R.A., and Spamer, E.E., America Bulletin, v. 1, p. 49–64. Wenrich, K.J., Huntoon, P.W., and Young, R.A., eds., eds., Colorado River Origin and Evolution: Grand Can- Wallace, M.A., Faulds, J.E., and Brady, R.J., 2005, Geo- Breccia-Pipe and Geologic Map of the Southwestern yon, Arizona, Grand Canyon Association, p. 215–222. logic Map of the Meadview North Quadrangle, Mo- Part of the Hualapai Indian Reservation and Vicinity, Spencer, J.E., Pearthree, P.A., and House, P.K., 2008a, have County, Arizona and Clark County, Nevada: Arizona: U.S. Geological Survey Map I-2554, Miscel- An evaluation of the evolution of the latest Miocene Nevada Bureau of Mines and Geology Map 154, scale laneous Investigations Series, p. 21–50, scale 1:48,000. to earliest Pliocene Bouse Lake system in the lower 1:24,000, 1 sheet. Young, R.A., 2001a, Geomorphic, structural and strati- Colorado River valley, southwestern USA, in Reheis, Wenrich, K.J., Billingsley, G.H., and Huntoon, P.W., 1996, graphic evidence for Laramide uplift of the southwest- M.C., Hershler, R., and Miller, D.M., eds., Late Ceno- Breccia-Pipe and Geologic Map of the Northeastern ern Colorado Plateau margin in northwest Arizona, in zoic Drainage History of the Southwestern Great Basin Part of the Hualapai Indian Reservation and Vicinity, Erskine, M.C., Faulds, J.E., Bartley, J.M., and Rowley, and Lower Colorado River Basin: Geologic and Biotic Arizona: U.S. Geological Survey Map I-2522, Miscel- P.D., eds., The Geologic Transition, High Plateaus Perspectives: Geological Society of America Special laneous Investigations Series, scale 1:48,000, 1 sheet. to Great Basin: A Symposium and Field Guide: The Paper 439, p. 375–390, doi: 10.1130/2008.2439(17). Wenrich, K.J., Billingsley, G.H., and Huntoon, P.W., 1997, Mackin Volume: Salt Lake City, Utah Geological Asso- Spencer, J.E., Smith, G.R., and Dowling, T.E., 2008b, Breccia-Pipe and Geologic Map of the Northwestern ciation, p. 227–237. Middle to late Cenozoic geology, hydrography and Part of the Hualapai Indian Reservation and Vicinity, Young, R.A., 2001b, The Laramide–Paleogene history of fi sh evolution in the American Southwest, in Reheis, Arizona: U.S. Geological Survey Map I-2440, Miscel- the western Grand Canyon region: Setting the stage, in M.C., Hershler, R., and Miller, D.M., eds., Late Ceno- laneous Investigations Series, scale 1:48,000, 1 sheet. Young, R.A., and Spamer, E.E., eds., Colorado River zoic Drainage History of the Southwestern Great Basin Wernicke, B., and Axen, G.J., 1988, On the role of Origin and Evolution: Grand Canyon, Arizona, Grand and Lower Colorado River Basin: Geologic and Biotic isostasy in the evolution of normal fault systems: Canyon Association, p. 7–16. Perspectives: Geological Society of America Special Geology, v. 16, p. 848–851, doi: 10.1130/0091-7613 Young, R.A., 2008, Pre–Colorado River drainage in west- Paper 439, p. 279–299. (1988)016<0848:OTROII>2.3.CO;2. ern Grand Canyon: Potential infl uence on Miocene Spencer, J.E., Richard, S.M., Gehrels, G.E., Gleason, J.D., Whistler, D.P., and Lander, E.B., 2003, New late Uintan to stratigraphy in Grand Wash Trough, in Reheis, M.C., and Dickinson, W.R., 2011, Age and tectonic setting of early Hemingfordian land mammal assemblages from Hershler, R., and Miller, D.M., eds., Late Cenozoic the Mesozoic McCoy Mountains Formation in western the undifferentiated Sespe and Vaqueros Formations, Drainage History of the Southwestern Great Basin Arizona, USA: Geological Society of America Bulletin Orange County, and from the Sespe and equivalent and Lower Colorado River Basin: Geologic and Biotic (in press). marine formations in Los Angeles, Santa Barbara, and Perspectives: Geological Society of America Special Spiegel, C., Kohn, B., Raza, A., Rainer, T., and Gleadow, Ventura Counties, Southern California: Bulletin of the Paper 439, p. 319–333. A., 2007, The effect of long-term low-temperature American Museum of Natural History, v. 279, p. 231– Young, R.A., and McKee, E.H., 1978, Early and Middle exposure on apatite fi ssion track stability: A natural 268, doi: 10.1206/0003-0090(2003)279<0231:C> Cenozoic drainage and erosion in west-central Ari- annealing experiment in the deep ocean: Geochimica 2.0.CO;2. zona: Geological Society of America Bulletin, v. 89, et Cosmochimica Acta, v. 71, p. 4512–4537, doi: Woodford, A.O., Welday, E.E., and Merriam, R., 1968, p. 1745–1750, doi: 10.1130/0016-7606(1978)89 10.1016/j.gca.2007.06.060. Siliceous tuff clasts in Upper Paleogene of south- <1745:EAMCDA>2.0.CO;2. Stewart, J.H., Gehrels, G.E., Barth, A.P., Christie-Blick, N., ern California: Geological Society of America Bul- Young, R.A., and Spamer, E.E., 2001, Colorado River Ori- and Wrucke, C.T., 2001, Detrital zircon provenance of letin, v. 79, p. 1461–1486, doi: 10.1130/0016-7606 gin and Evolution: Grand Canyon, Arizona, Grand Mesoproterozoic to Cambrian arenites in the western (1968)79[1461:STCITU]2.0.CO;2. Canyon Association, 280 p. United States and northwestern Mexico: Geological Young, R.A., 1966, Cenozoic Geology along the Edge of Zachos, J., Pagani, M., Sloan, L., Thomas, E., and Billups, Society of America Bulletin, v. 113, p. 1343–1356, the Colorado Plateau in Northwestern Arizona [Ph.D. K., 2001, Trends, rhythms, and aberrations in global doi: 10.1130/0016-7606(2001)113<1343:DZPOMT> thesis]: St. Louis, Washington University, 167 p. climate 65 Ma to present: Science, v. 292, p. 686–693, 2.0.CO;2. Young, R.A., 1979, Laramide deformation, erosion and doi: 10.1126/science.1059412. Stöckli, D.F., 2005, Thermochronometric constraints on plutonism along the southwestern margin of the Colo- Zeilinger, G., and Schlunegger, F., 2007, Possible fl ex- Tertiary erosional exhumation and incision of the rado Plateau: Tectonophysics, v. 61, p. 25–47, doi: ural accommodation on the eastern edge of the Alti- Colorado Plateau—A progress report: Geological So- 10.1016/0040-1951(79)90290-7. plano in relation to focused erosion in the Rio La Paz ciety of America Abstracts with Programs, v. 37, no. 7, Young, R.A., 1982, Paleogeomorphologic evidence for the drainage system: Terra Nova, v. 19, p. 373–380, doi: p. 109. structural history of the Colorado Plateau margin in 10.1111/j.1365-3121.2007.00762.x. Stüwe, K., White, L., and Brown, R., 1994, The infl u- western Arizona, in Frost, E.G., and Martin, D.L., eds., ence of eroding topography on steady-state iso- Mesozoic-Cenozoic Tectonic Evolution of the Colo- therms—Application to fi ssion-track analysis: Earth rado River Region, California, Arizona and Nevada MANUSCRIPT RECEIVED 16 FEBRUARY 2010 and Planetary Science Letters, v. 124, p. 63–74, doi: (Anderson-Hamilton Volume): San Diego, California, REVISED MANUSCRIPT RECEIVED 6 MAY 2010 MANUSCRIPT ACCEPTED 28 JUNE 2010 10.1016/0012-821X(94)00068-9. Cordilleran Publishers, p. 29–39. Topping, D.J., Rubin, D.M., and Vierra, L.E., 2000, Colo- Young, R.A., 1985, Geomorphic evolution of the Colorado rado River sediment transport: 1. Natural sediment Plateau margin in west-central Arizona: A tectonic Printed in the USA

Geological Society of America Bulletin, Month/Month 2010 29