GSA Bulletin; January 2008; v. 120; no. 1-2;
p. 13-33; DOI: 10.1130/B26081.1
© 2008 Geological Society of America
The Chuska erg: Paleogeomorphic and paleoclimatic implications of an Oligocene sand sea on the Colorado Plateau
Steven M. Cather
,1,
Sean D. Connell1,
Richard M. Chamberlin1,
William C. McIntosh1,
Glen E. Jones1,
Andre R. Potochnik2,
Spencer G. Lucas3 and
Peggy S. Johnson4
1 New Mexico Bureau of Geology and Mineral Resources, New Mexico Institute of Mining and Technology, 801 Leroy Place, Socorro, New Mexico 87801, USA
2 18 E. Juniper Avenue, Flagstaff, Arizona 86001, USA
3 New Mexico Museum of Natural History, 1801 Mountain Road NW, Albuquerque, New Mexico 87104, USA
4 New Mexico Bureau of Geology and Mineral Resources, New Mexico Institute of Mining and Technology, 801 Leroy Place, Socorro, New Mexico 87801, USA
 |
ABSTRACT
|
|---|
Great thicknesses of eolian dune deposits of early Oligocene age crop out in the Chuska Mountains of northwestern New Mexico-Arizona (as much as 535 m thick) and in the Mogollon-Datil volcanic field of western New Mexico-Arizona (as much as 300 m thick). 40Ar/39Ar ages of intercalated volcanic rocks indicate eolian deposition in these areas was approximately synchronous, with eolian accumulation beginning regionally at ca. 33.5 Ma and ending at ca. 27 Ma. Probable eolian sandstone of Oligocene age 483 m thick is also present in the subsurface of the Albuquerque Basin of the Rio Grande rift. The beginning of eolian deposition on the Colorado Plateau corresponds closely to the beginning of eolian (loessic) deposition in the White River Group of the Great Plains and major Oi1 glaciation in Antarctica, suggesting possible global paleoclimatic control.
Successions of Oligocene eolian sandstone on the Colorado Plateau are thicker than all of the better known Upper Paleozoic-Mesozoic eolianites in the region, except the Jurassic Navajo Sandstone. We suggest that the widely separated Oligocene eolianites in the Colorado Plateau region were probably originally continuous, and thus are erosional remnants of an extensive (
140,000 km2), regional sand sea (the Chuska erg). This interpretation is based on: (1) comparison with thickness trends of older eolianites in the Colorado Plateau region, (2) evaluation of regional topographic gradients of modern ergs, and (3) hydrologic modeling of a 300- to 400-m–thick zone of saturation that existed during eolian deposition in the Chuska Mountains.
The Chuska erg represents the final episode of Paleogene aggradation on the central and southern Colorado Plateau. Aggradation was driven primarily by trapping of fluvial sediments on the plateau by development of major volcanic fields along the eastern plateau margin. These volcanic fields blocked earlier Laramide drainages that had previously transported sediments eastward off the plateau. Following a shift to widespread eolian deposition at ca. 33.5 Ma, constructional volcanic topography induced eolian accumulation upwind of developing volcanic fields. Stratal accumulation rates (not decompacted) of eolian deposits were 28–82 m/m.y.
The reconstructed top of the Chuska erg would lie at a present-day elevation of 3000 m or more, and provides a datum for assessing subsequent erosion on the Colorado Plateau. Major exhumation (
1230 m) occurred during the late Oligocene and early Miocene, following the end of Chuska deposition and prior to the onset of Bidahochi Formation deposition at ca. 16 Ma on the south-central part of the plateau. The Bidahochi Formation attained a thickness of 250 m by ca. 6 Ma, followed by 520 m of late Miocene and younger erosion in the valley of the Little Colorado River. The depth of late Oligocene-early Miocene (ca. 26–16 Ma) exhumation of the central and southern Colorado Plateau thus was more than twice that of the late Miocene-Holocene (ca. 6–0 Ma). The timing of initial deep erosion in the Colorado Plateau-Southern Rocky Mountains region suggests the beginning of major epeirogenic rock uplift occurred during post-Laramide magmatism.
Key Words: eolian • erg • paleoclimate • Colorado Plateau • Oligocene • epeirogeny • exhumation
|
INTRODUCTION
|
|---|
The deposits of eolian sand seas (ergs) document regional episodes of aridity in the geologic record. Deposits of ancient ergs, however, are commonly poorly preserved due to effects of subsequent tectonism and erosion. Reconstruction of the original geometry of incompletely preserved ancient erg deposits commonly relies on qualitative interpretation. Paleoclimatic events that contribute to erg development are often inferential.
In the Colorado Plateau region of southwestern North America, thick eolian sandstones of Oligocene age occur in widely scattered localities in New Mexico and Arizona. In this paper, we present stratigraphic, radioisotopic, paleohydrologic, and paleogeomorphic data that suggest that these deposits were part of an extensive eolian sand sea (hereafter termed the Chuska erg) that occupied much or all of the central and southern parts of the Colorado Plateau from ca. 33.5 Ma to ca. 27 Ma. Possible global, paleoclimatic controls for erg development and implications for the Cenozoic uplift and exhumation of the Colorado Plateau are also discussed.
We focus on the regional stratigraphic framework of lower Oligocene, eolian dune deposits exposed in the Chuska Mountains and Mogollon-Datil volcanic field of New Mexico and Arizona and sandstone of probable eolian origin in the subsurface of the Albuquerque Basin of the Rio Grande rift (Fig. 1). The presence of thick eolianites in the Chuska Mountains was first noted by Wright (1954, 1956) and in the western Mogollon-Datilfield by Wrucke(1961),whowas also the first to propose the possible correlation of the Chuska and Mogollon-Datil eolianites. These eolianites were deposited during the transition between the better known Laramide (Late Cretaceous-Eocene) and rift-related (Miocene) episodes of the geologic history of the Colorado Plateau. Rocks of Oligocene age are only locally preserved on the Colorado Plateau, leaving large regions of the plateau with an incomplete Cenozoic stratigraphic record.
View larger version (43K):
[in this window]
[in a new window]
|
Figure 1. Map showing Oligocene eolian sandstone localities and Oligocene volcanic fields and intrusions in relation to relict Laramide basins, uplifts, and the schematic Laramide paleodrainage net for the Colorado Plateau (after Potochnik, 2001b). Selected Laramide uplifts are: Du—Defiance uplift; Zu—Zuni uplift; Ku—Kaibab uplift; Mu—Monument uplift; SJu—San Juan uplift; Uu—Uinta uplift. Laramide basins are: Ub—Uinta Basin; GRb—greater Green River Basin; Pb—Piceance Basin; NPb—North Park-Middle Park Basin; Db—Denver Basin; Cb—Claron Basin; SPb—South Park Basin; Rb—Raton Basin; SJs—San Juan sag; Gb—Galisteo-El Rito Basin; CJb—Carthage-La Joya Basin; Sb—Sierra Blanca Basin.
|
|
|
EARLY OLIGOCENE EOLIANITES IN THE COLORADO PLATEAU AREA
|
|---|
Chuska Mountains
The Chuska Mountains lie near the center of the Colorado Plateau (Fig. 2) and contain a thick succession of sandstone that unconformably overlies Mesozoic strata (Gregory, 1917; Darton, 1928; Wright, 1954, 1956). This succession, termed the Chuska Sandstone by Gregory (1917), records the culminant episode of Paleogene sedimentation on the central Colorado Plateau and, with the exception of laccolith-related highlands in areas to the north, occupies the highest elevations on the central plateau. Understanding the timing and genesis of the Chuska Sandstone thus is crucial to the interpretation of the stratigraphic and paleogeomorphic development of the Colorado Plateau.
View larger version (56K):
[in this window]
[in a new window]
|
Figure 2. (A) Digital elevation model for the Chuska Mountains area. Flat top of the southern Chuska Mountains is defined by silica-cemented zones within the Chuska Sandstone that formed during early, phreatic diagenesis (see text). Note the eastern boundary structure of the Laramide Defiance uplift (the Defiance monocline, DM) is buried by the Chuska Sandstone. Selected features of the Navajo volcanic field are Buell Park diatreme (BP), Narbona Pass maar (NP), sill at Beautiful Mountain (BM), and the Hidden Valley maar (HV). Volcanic necks at Roof Butte (RB) mark highest part of range (2980 m). (B) Simplified geologic map of the Chuska Mountains area, after Wright (1956).
|
|
The Chuska Sandstone consists of a basal fluvial unit 0–81 m thick (the Deza Member) that is transitionally overlain by eolian dune deposits (Narbona Pass Member) as much as 535 m thick in the northern Chuska Mountains (Lucas and Cather, 2003; Cather et al., 2003). The Chuska Sandstone is flat lying and undeformed, and overlies a low-relief paleoerosion surface 2350–2440 m in elevation that beveled folded Mesozoic strata along the east flank of the Laramide Defiance uplift (Fig. 2; Cather et al., 2003). The paleoerosion surface is reasonably well exposed for 80 km along the flanks of the Chuska Mountains, except in the southeast part of the range, where it is obscured by extensive landslides. The paleoerosion surface is gently undulatory with broad, shallow paleovalleys that are typically less than a few tens of meters in depth. The basal fluvial part of the Chuska Sandstone onlapped parts of this low-relief, post-Laramide surface (Tsaile surface of Cooley, 1958, and Schmidt, 1991) in response to the south-southwesterly progradation of the distal part of an extensive (200-km length) piedmont-drainage system that headed in the San Juan Mountains area of southwestern Colorado (Cather et al., 2003). Eolian crossbeds in the overlying Narbona Pass Member (Fig. 3) are as much as 6 m high and record paleowinds mostly from the south and west (Wright, 1956; Trevena, 1979; Repenning et al., 1958). Eolian sediment transportation was generally up the regional paleoslope established during Deza fluvial deposition. Inversely graded translatent-ripple laminae, grain-flow laminae, and grain-fall laminae are common; these are characteristic of eolian sedimentation (Hunter, 1981).
View larger version (156K):
[in this window]
[in a new window]
|
Figure 3. Eolian crossbeds in Chuska Sandstone (Narbona Pass Member) at 2950 m elevation near Roof Butte, northern Chuska Mountains. Hammer for scale at center of photograph.
|
|
The Chuska Sandstone contains arenaceous detritus derived from plutonic, volcanic, recycled sedimentary, and metamorphic sources (QFL (quartz feldspar lithics) 55:35:10; see Wright, 1956; Trevena, 1979; Trevena and Nash, 1981; and Cather et al., 2003; for provenance details). Petrographic similarity between the Deza and Narbona Pass members and lack of major upsection changes in detrital composition suggest similar sources for each unit. Both members may consist of windblown sand derived primarily from the southwest. In the Deza Member, these eolian sands were subsequently fluvially reworked, although minor eolian strata are locally preserved. Fluvial systems that drained the relict Laramide San Juan uplift and the nascent San Juan volcanic field also contributed fluvial detritus to the Deza Member (Cather et al., 2003).
Narbona Pass Member eolianites are locally well cemented with early diagenetic opal and chalcedony (Fig. 4). Detrital grains in these sandstones typically exhibit uniform (isopachous) rims of siliceous cements that are characteristic of phreatic diagenesis. Pendulose and meniscate cement fabrics associated with vadose diagenesis are rare (Cather et al., 2003). Silica-cemented beds of eolian sandstone are intercalated with friable, weakly cemented eolian sandstone throughout the Chuska Mountains and give rise to ledgy outcrops. Phreatically cemented horizons define the flat range top that characterizes the southern two-thirds of the Chuska Mountains at elevations of 2730–2850 m (Fig. 2). Higher elevations in the northern part of the range are held up mostly by volcanic necks and sills that intrude the Chuska Sandstone. Silica-cemented horizons in the Chuska Mountains represent paleogroundwater tables that were at least 300–400 m above the base of the Chuska Sandstone. The high precement porosity of these rocks (26.2%–42.1%, mean 33.2%) indicates that phreatic cementation occurred before significant compaction and implies that the water table rose to keep pace with eolian aggradation (Cather et al., 2003). Accumulation of eolianites by the rising capillary fringe of very shallow groundwater during Chuska sedimentation, however, does not appear to have occurred. Such sedimentation is characterized by a dominance of interdune deposits (e.g., Kocurek, 1996), which is not seen in the Chuska Sandstone. Closely following the end of Chuska eolianite deposition, interaction between shallow groundwater and ascending magmas associated with the Navajo volcanic field produced several phreatomagmatic eruption centers (maars) in the Chuska Mountains area (Semken, 2003).
View larger version (199K):
[in this window]
[in a new window]
|
Figure 4. Thin section of eolian sandstone of Narbona Pass Member of the Chuska Sandstone showing isopachous rims of chalcedony cement, indicative of cementation in a phreatic environment. Width of field is 0.5 mm.
|
|
Early workers suggested a variety of ages for the Chuska Sandstone that ranged from Eocene to Pliocene, based mostly on presumed correlation with other units on the Colorado Plateau or from paleogeographic considerations (see summary in Lucas and Cather, 2003). Cather et al. (2003) provided the first direct ages from the Chuska Sandstone, using the 40Ar/39Ar technique [note that ages produced by the New Mexico Geochronology Research Laboratory at New Mexico Tech have been recalibrated in this paper using the monitor calibration of Renne et al. (1998; Fish Canyon Tuff sanidine = 28.02 Ma)]. These include: (1) a 34.97 ± 0.20-Ma age (all errors ±2
) from a fall-out ash in the middle part of the fluvial Deza Member in the southern end of the Chuska Mountains; (2) a 33.54 ± 0.25-Ma age from a fall-out ash in the lower part of the eolian Narbona Pass Member in the northern part of the range; and (3) ages of 24.83 ± 0.26, 25.13 ± 0.16, 25.21 ± 0.17, and 25.40 ± 0.17 Ma (weighted mean average 25.21 ± 0.16 Ma) for Navajo volcanic field trachybasaltic lavas that disconformably overlie eolianites of the Chuska Sandstone at Narbona Pass.
Radioisotopic data indicate eolian deposition in the central Colorado Plateau began between 34.97 ± 0.20 and 33.54 ± 0.25 Ma and ended before 25.21 ± 0.16 Ma (Fig. 5A). The timing of the termination of eolian deposition in the Chuska Mountains is not well defined. Significant erosional paleorelief (Gregory, 1917; Wright, 1956; Lucas and Cather, 2003) developed at the top of the Chuska Sandstone prior to eruption of overlying trachybasalts at 25.21 ± 0.16 Ma.
View larger version (43K):
[in this window]
[in a new window]
|
Figure 5. Chronostratigraphic correlation diagram for eolianites and associated rocks in the Chuska Mountains and the Mogollon-Datil volcanic field. (A) Chuska Mountains, showing members of Chuska Sandstone. (B) Western Mogollon-Datil field. (C) Northern Mogollon-Datil field. See text for sources of radioisotopic age data. Tmv—lava flows associated with maar at Narbona Pass; Tbp—Bishop Peak Tuff; Tcl—volcaniclastic unit of Cañon del Leon; Ttl—tuff of Luna; Tcb—Caballo Blanco Tuff; Tdc—Davis Canyon Tuff; Tvp—Vicks Peak Tuff; Tss—Squirrel Springs Canyon Andesite; Tsp—Shelly Peak Tuff; Tbg—Bloodgood Canyon Tuff; Tdw—Datil Well Tuff; Trh—Rock House Canyon Tuff; Tbc—Blue Canyon Tuff; Thm—Hells Mesa Tuff; Tcp—South Crosby Peak Formation; Tlj—La Jencia Tuff; Tlp—La Jara Peak Basaltic Andesite; Ths—basaltic andesite of Hidden Spring; Tsc—South Canyon Tuff; Tlm—Lion Mountain Andesite; Tcba—basaltic andesite of Crosby Mountains; Tts—tuff of Turkey Springs. Numbers in parentheses indicate radioisotopic age (Ma).
|
|
The Chuska Sandstone is preserved at elevations as high as 2980 m near Roof Butte in the northern Chuska Mountains, but the original depositional top of the Chuska Sandstone is not preserved there or in any other part of the range. Thus, the 535-m thickness of Chuska eolianite present in the northern part of the range is a minimum estimate of the original thickness.
Mogollon-Datil Volcanic Field
Oligocene eolian sandstone is exposed in the northwest part of the Mogollon-Datil volcanic field (Fig. 6). As much as 200–300 m thick in the western part of the field, these eolianites thin to the south and east as they grade laterally into volcanic and noneolian, volcaniclastic rocks toward the center of the volcanic field. Eolian sandstones are mostly fine to medium grained, well sorted, and exhibit large-scale cross bedding as much as 10 m high (Chamberlin and Harris, 1994). These sandstones are in the upper part of the upper Eocene-Oligocene Spears Group (sensu Cather et al., 1994), but have been accorded various formal and informal names in different parts of the volcanic field (Table 1).
View larger version (34K):
[in this window]
[in a new window]
|
Figure 6. Simplified geologic map of the northern Mogollon-Datil volcanic field showing thickness distribution of Oligocene eolian sandstone. Thickness data from Wrucke (1961), Chamberlin and Harris (1994), Harrison (1980), Bornhorst (1976), Lopez (1975), Cather (1986), Coffin (1981), Ferguson (1986), Ratté (1980, 1981, 1989), and Lawrence and Richter (1986). Circled numerals correspond to wells: (1) Belcher 1/State; (2) Alpine 1/Federal. Hfs—Hickman fault system; RLfs—Red Lake fault system; Ab—Albuquerque Basin; Sab—San Agustin Basin; Rg—Reserve graben; Bg—Blue graben; Dm—Datil Mountains; Gm—Gallinas Mountains.
|
|
Because eolian sandstone of the northwestern Mogollon-Datil volcanic field interfingers with radioisotopically dated volcanic rocks (Fig. 7), the timing of eolian sedimentation is well defined (Table 2; Fig. 5). The youngest dated unit that underlies the eolian succession is the tuff of Luna (33.51 ± 0.14 Ma), which crops out locally north and west of Reserve, New Mexico. Eolian sandstone in the Reserve-Alpine-Quemado area in the western part of the Mogollon-Datil field locally interfingers with six dated ignimbrites (32.26–28.23 Ma). The oldest dated rock overlying the eolian succession is the Bearwallow Mountain Andesite (ca. 26.3– 25 Ma). If eolian sedimentation continued in the western Mogollon-Datil field after the eruption of the Bearwallow Mountain Andesite, no record of it is preserved.
View larger version (151K):
[in this window]
[in a new window]
|
Figure 7. Lenticular, nonwelded ignimbrite (Tvp—Vicks Peak Tuff; 28.8 Ma) inter-bedded with eolian sandstone near Mangas, New Mexico. Note only minor erosion of the avalanche face of dune beneath ignimbrite.
|
|
View this table:
[in this window]
[in a new window]
|
TABLE 2. RADIOISOTOPIC AGES OF SELECTED VOLCANIC ROCKS ASSOCIATED WITH EOLIANITES IN THE WESTERN MOGOLLON-DATIL VOLCANIC FIELD
|
|
Throughout the western part of the Mogollon-Datil volcanic field, eolianites transitionally overlie fluvial volcaniclastic rocks of the middle Spears Group that were derived from the south and east (eolian thicknesses reported in this paper locally include significant thicknesses of intercalated fluvial deposits in the lower part of the eolian succession). The upper contact of the eolian deposits in this region is disconformable with the overlying Bearwallow Mountain Andesite. The upper contact exhibits modest paleorelief (a few tens of meters or less), except near Alpine, Arizona, where 200 m of paleorelief may be present (Wrucke, 1961, p. 22). These contact relationships suggest significant erosion occurred before the ca. 26.3- to 25-Ma eruption of the Bearwallow Mountain Andesite. The lacuna associated with this unconformity, however, cannot be longer than ca. 2–3 m.y., because ignimbrites as young as 28.2 Ma are interbedded with eolianites below the contact (Fig. 5B).
In the Gallinas and Datil Mountains of the northern Mogollon-Datil field, eolianites are thin (<50 m thick; Fig. 6) and were deposited over a shorter time interval than in areas to the west (Fig. 5C). This thinning is due largely to interfingering with more proximal volcanic and noneolian, volcaniclastic rocks to the east. Eolian deposits also display a seemingly abrupt thinning eastward across the Hickman fault system (Hfs; Fig. 6). This fault system was an active west-down structure beginning ca. 36 Ma (Cather, 1990) and appears to have influenced the thickness of eolian deposits.
Thick (>100-m) eolianites also comprise part of the fill of the Mount Withington caldera in the San Mateo Mountains (Ferguson, 1986). These eolianites consist of dune deposits that exhibit cross bedding as much as 7 m high and record paleowinds from the southwest (Ferguson, 1986). The eolian strata lie stratigraphically between the 27.6-Ma South Canyon Tuff and the 24.5-Ma Turkey Springs Tuff. These deposits are the easternmost occurrence of thick eolianites in the Mogollon-Datil volcanic field.
The petrology of eolian sandstone in the Mogollon-Datil volcanic field has only been studied at a few localities. In general, eolianites consist largely of volcaniclastic detritus that presumably was derived from within the field (Wrucke, 1961; Ferguson, 1986; Chamberlin and Harris, 1994). Minor nonvolcanic material (principally microcline grains) reflects basement derivation from unknown sources. Heulandite is the principal cement; less abundant is calcite (Wrucke, 1961; Chamberlin and Harris, 1994).
Subsurface of Northwestern Albuquerque Basin
The Albuquerque Basin (Fig. 8) is one of the largest extensional basins of the Rio Grande rift and contains a 5-km–thick fill of the upper Oligocene-Pleistocene Santa Fe Group (Connell, 2004). Underlying the Santa Fe Group is the unit of Isleta #2 (Lozinsky, 1994), a 483- to 2185-m succession of Paleogene sandstone, mudstone, and volcanic rocks found only in deep drillholes.
View larger version (61K):
[in this window]
[in a new window]
|
Figure 8. Digital elevation model for the Albuquerque Basin area, showing basin boundaries (dashed lines) and cross-section line (Fig. 10). Wells are: (1) Tamara #1-Y; (2) West Mesa Federal #1; (3) Shell Isleta #2; (4) TransOcean Isleta #1.
|
|
We interpret that at least one well in the northwestern Albuquerque Basin penetrated an eolian sandstone facies within the unit of Isleta #2. The Tamara #1-Y well encountered 483 m of mostly well-sorted sandstone and minor mudstone in the unit of Isleta #2 (Connell et al., 2001). The upper 247 m contains pink to very pale-brown, mostly fine- to medium-grained, quartz-rich feldspathic arenite and lithic arkose with scattered traces of white ash and sparse purplish-gray volcanic lithic grains (Connell et al., 2001; Koning and Personius, 2002). The lower 236 m contains locally cemented, reddish-yellow, fine- to coarse-grained subarkose and lithic arkose with rounded to subrounded grains. Lithic fragments are mostly chert and gray volcanic grains. Petrographic examination of sandstone from four depth intervals within the unit of Isleta #2 in the Tamara #1-Y indicates subarkosic compositions (QFL 69:19:12) with common volcanic lithic grains (Connell et al., 2001). The lower 55 m of the unit contains more volcanic grains than the upper part.
The unit of Isleta #2 in the Tamara #1-Y well is suggestive of eolian deposition for several reasons. First, cuttings show the sand is generally well sorted, and grains are typically rounded to subrounded and exhibit frosted surfaces. Second, borehole geophysical logs for the Tamara #1-Y well show an overall smooth to blocky electrical resistivity log signature (Fig. 9). The relatively uniform geophysical characteristics of the unit of Isleta #2 in the Tamara #1-Y well suggest that it consists mostly of homogeneous sandstone. The log signature differs from that of known fluvial deposits in the well, but is similar to the overlying eolian parts of the Zia Formation and to eolian sandstones in other areas (e.g., Lupe and Ahlbrandt, 1979; Vincelette and Chittum, 1981). Unfortunately, definitive criteria for the discrimination of eolian deposits in the Tamara #1-Y well, such as dip-meter data and cored intervals, are unavailable. Thin intervals of mudstone in the unit of Isleta #2 suggest fluvial or lacustrine deposits locally intertongue with eolianites.
View larger version (18K):
[in this window]
[in a new window]
|
Figure 9. Gamma-ray and conductivity logs for Tamara #1-Y well. The unit of Isleta #2 is interpreted to be largely eolian in this well. amsl—above mean sea level; mmhos—millimhos (= 10–3 siemens).
|
|
The thickness of the probable eolianite in the Tamara #1-Y well (483 m) is comparable to Oligocene eolianite thicknesses in the Chuska Mountains (535 m) and the western Mogollon-Datil volcanic field (200–300 m). The eolian facies present in the Tamara #1-Y well probably interfingered with contemporaneous volcanic and noneolian, volcaniclastic and siliciclastic rocks of the unit of Isleta #2 to the north, east, and south. In the southern Albuquerque Basin, these volcanic and volcaniclastic rocks were derived from the Mogollon-Datil volcanic field, and thus were continuous with the northern volcaniclastic apron of the field. To the east, the eolianite may have been contiguous with volcaniclastic strata of the upper Eocene-Oligocene Espinaso Formation.
Geochronologic control is limited for the unit of Isleta #2. The only direct age constraint is a K-Ar date of 36.3 ± 1.8 Ma reported for an ignimbrite in the Shell Isleta #2 well (Fig. 10; May and Russell, 1994) from a fluvial part of the unit. The unit of Isleta #2 is overlain by eolian sandstone of the Piedra Parada Member of the Zia Formation (ca. 22–19 Ma; Tedford and Barghoorn, 1999). Additional age constraints come from the northwestern margin of the Albuquerque Basin, where the unit of Isleta #2 was eroded prior to deposition of the Zia Formation. There, the Zia Formation unconformably overlies the Eocene Galisteo Formation and Upper Cretaceous strata. Locally, this contact is marked by a scattered lag of volcanic pebbles and cobbles that have been sculpted into ventifacts (Tedford and Barghoorn, 1999). 40Ar/39Ar analyses of hornblende and biotite from three of these ventifacts establish an early Oligocene maximum age (32.0 ± 1.4 Ma, 33.45 ± 0.24 Ma, 33.24 ± 0.22 Ma) for this clast layer. Taken together, age constraints in the Albuquerque Basin area indicate that the unit of Isleta #2 was deposited between late Eocene and earliest Miocene time, ca. 37–22 Ma.
View larger version (18K):
[in this window]
[in a new window]
|
Figure 10. Cross-section A–A' of Cenozoic rocks in the northern and central Albuquerque Basin, showing interpreted relationships between Oligocene eolian and non-eolian strata. Cross-section line is shown in Figure 8. Modified from Connell et al. (2001).
|
|
|
PALEOGEOGRAPHIC EXTENT OF OLIGOCENE EOLIANITES
|
|---|
Oligocene eolian sandstone localities in the Chuska Mountains, the Mogollon-Datil volcanic field, and the Albuquerque Basin are separated from each other by 200 km, and, in the area between them, Oligocene rocks have been removed by erosion. Do these eolian deposits represent local dune fields, or are they erosional remnants of an erg of regional extent? In the following sections, we attempt to address this question by: (1) evaluating the stratigraphic thinning rate of better preserved, Upper Pennsylvanian-Jurassic eolianites in the Colorado Plateau area; (2) examining the regional surface slopes of modern ergs; and (3) calculating the minimum lateral extent of eolian sand necessary to support a saturated groundwater zone 300–400 m thick, given representative hydraulic conductivities for eolian sand and rates of precipitation in modern areas of eolian deposition.
Eolianites in the Chuska Mountains were at least 535 m thick, thicker than any of the better known, Upper Paleozoic-Mesozoic eolian sandstones of the Colorado Plateau area, except the Jurassic Navajo Sandstone (660 m thick). Many of the Upper Paleozoic-Jurassic eolianites of the Colorado Plateau were thickened in local areas of syndepositional subsidence (Blakey et al., 1988; Blakey, 1988). The eolianites in the Chuska Mountains, however, were clearly not deposited in a local basin, as shown by the undeformed, flat-lying basal contact and by the fact that the Chuska Sandstone overlies the structurally high Defiance uplift and laps across its beveled monoclinal margin without deformation (Wright, 1956; Cooley, 1958: Schmidt, 1991; Lucas and Cather, 2003; Cather et al., 2003). The lack of significant post-Laramide paleotopography near the Chuska Mountains also implies that the Chuska Sandstone did not accumulate adjacent to a local highland, as have the Great Sand Dunes of Colorado. The factors that control erg growth and preservation will be discussed in a later section.
It is possible to broadly estimate the former lateral extent of the Chuska Sandstone by examining stratigraphic thinning rates of other, better preserved, ancient eolian sandstones. Using the isopach maps of Blakey et al. (1988), we calculated the maximum rate of lateral thinning of parts of ten Upper Pennsylvanian-Middle Jurassic erg deposits in the Colorado Plateau region. We limited our analysis to areas in which the isopach interval contains >50% eolian sandstone (Blakey et al., 1988). To ensure a conservative comparison, we tabulated data only from those areas in which the most rapid lateral thinning for each isopach interval occurs. We did not attempt to correct for the effect of post-depositional erosion of the margins of erg deposits, an effect that tends to increase the rate of lateral thinning. Maximum thinning rates for these erg deposits range between 0.3 and 2.5 m/km, averaging 1.4 m/km (Table 3). If a rapid thinning rate (2.5 m/km) is assumed, then the former margin of the Chuska Sandstone was 200 km from the Chuska Mountains. This line of reasoning implies that the isolated Oligocene eolianites described in this report are remnants of a regionally extensive erg.
View this table:
[in this window]
[in a new window]
|
TABLE 3. MAXIMUM STRATIGRAPHIC THINNING RATES OF SELECTED ERG DEPOSITS OF LATE PALEOZOIC AND MESOZOIC AGE IN THE COLORADO PLATEAU REGION
|
|
Preserved remnants of post-Laramide, paleogeomorphic surfaces on the central Colorado Plateau (Schmidt, 1991; Cather et al., 2003) and in the southern Rocky Mountains (Epis and Chapin, 1975) indicate that relict Laramide paleotopography had been beveled to low relief in these areas by the late middle Eocene (see also Pederson et al., 2002). If it is assumed that these low-relief remnants are representative of the regional surface upon which the Chuska Sandstone was deposited, then the slope of the upper surface of the Chuska eolianite may have been a primary factor in determining the original lateral extent of the eolian sand body. Because the depositional top of the Chuska Sandstone is not preserved, we use the regional surface slopes of modern ergs as proxies.
We examined the slope characteristics of twelve large, presently active ergs using a Geographic Information system [Environmental Systems Research Institute (ESRI)]. Digital elevation models (DEMs) with a resolution of three arc-seconds (90 m) were downloaded from ftp://e0srp01u.ecs.nasa.gov/srtm/, using Version 2 SRTM (Shuttle Radar Topography Mission) data. The National Geospatial Intelligence Agency processed these data to remove single pixel errors and applied a vectorized coastline mask. One-degree tiles were assembled to form a single DEM for each major erg. Topographic contours were created using ESRI's Spatial Analyst Extension and were overlain with the WSI-Earth99-2k image from World-Sat International and ESRI. Contours appear as broad bands in the resulting maps (Fig. 11) due to the effects of dune topography. Minimum, maximum, and average regional slopes were estimated for each erg using conventional slope (rise/run) calculations. Slopes were analyzed only in areas where eolian sand cover is relatively complete (as determined from satellite imagery), to minimize the effect of pre-erg topography on erg slopes.
View larger version (109K):
[in this window]
[in a new window]
|
Figure 11. Topography of selected modern ergs. Contour interval is 100 m; elevations are relative to mean sea level. Dotted white lines delimit areas of relatively complete eolian sand cover. (A) Grand Erg Oriental, Algeria and Tunisia; (B) Great Sand Sea, Egypt and Libya; (C) Grand Erg Occidental, Algeria; (D) Aoukâr, Makteïr, and Ouarane ergs, Mauritania; (E) northwestern Simpson Desert, Australia; (F) Peski Kyzyl-Kum and Peski Kara-Kum, Uzbekistan and Turkmenistan; (G) Erg Bilma-Tènèrè, Niger; (H) Takla Makan, People's Republic of China; (I) Rub'al Khali, Saudi Arabia. Slope data for ergs are presented in Table 4.
|
|
The average regional surface slope of the studied ergs is 0.8 m/km, and ranges between 0.3 and 1.3 m/km (Table 4). Maximum slopes average 1.6 m/km, and typically occur where erg margins have onlapped adjacent highlands. For example, the steepest erg slope encountered in this study (3.3 m/km) occurs in the northwestern part of the Erg Bilma-Ténéré, Niger, but persists for only a short distance (100 km; Fig. 11G), where the erg has onlapped an adjacent highland.
These erg-slope data imply that the now-isolated Oligocene eolianites in the Colorado Plateau area were probably originally continuous. This is perhaps best illustrated by the 500-m–thick eolianites in the Chuska Mountains and the northwestern Albuquerque Basin, which are separated by 200 km, mostly along the structurally uniform Chaco homocline south of the San Juan Basin (Fig. 1). For these eolianites to have pinched out at the midpoint between them as a result of erg-surface topography would require regional slopes of 5 m/km, significantly steeper than observed in modern ergs.
Paleohydrologic data can also be used to model the minimum lateral extent of the Chuska Sandstone. Petrographic evidence described previously indicates the paleogroundwater table was at least 300–400 m above the base of the Chuska Sandstone. If the Chuska eolianite accumulated in a local dune field with a convex upper surface (Fig. 12), the lateral dimensions of the dune field needed to support such a phreatic system can be calculated. We assume that: (1) the base of the eolian sequence is a low-relief surface that overlies relatively impermeable Mesozoic strata; (2) the hydraulic conductivity of the eolianite is 10–5 m/s (Hong et al., 1999); and (3) recharge to the aquifer is 15 cm/yr, based on the maximum rainfall typical of modern ergs (Wilson, 1973), and we assume no losses to evapotranspiration. We calculate the radius of a dune field necessary to support a 300-m–thick saturated zone to be 640 km, based on Darcy flow analysis. A dune field of these dimensions is implausible because it would extend beyond the margins of the Colorado Plateau in all directions. The thick, paleosaturated zone present in the Chuska Sandstone thus is not compatible with a paleohydrologic system developed within a local dune field, but rather with one that developed within a regional eolian aquifer. Because rainfall rates typical for ergs cannot account for the evidence of a thick, paleosaturated zone, the eolian system must have developed within a low-lying basin where it could be recharged by adjacent fluvial systems. Given the fact that the Chuska Sandstone rests atop a Laramide uplift, the scale of such a basin would have been much larger than the present extent of the Chuska Mountains.
View larger version (12K):
[in this window]
[in a new window]
|
Figure 12. Hydrologic model to predict the lateral extent of a convex-up dune field necessary to support a phreatic groundwater system 300 m thick. See text for discussion.
|
|
In summary, we conclude that the eolian strata preserved in widely separated localities in the Chuska Mountains, the Mogollon-Datil volcanic field, and in the western Albuquerque Basin were probably part of an originally continuous erg. If a rapid, lateral thinning rate of 2.5 m/km is assumed, the erg margins were 200 km from the Chuska Mountains (Fig. 13). The erg thus occupied much or all of the low-lying area between the Mogollon Highland and the volcanic highlands that flanked the Colorado Plateau on the east. The southeast margin of the erg is well preserved in the Mogollon-Datil volcanic field, and the eastern margin is reasonably well located in the subsurface of the Albuquerque Basin. Elsewhere, the locations of former margins of the Chuska erg are imprecisely defined.
View larger version (40K):
[in this window]
[in a new window]
|
Figure 13. Map showing interpreted early Oligocene paleogeography in the Colorado Plateau-Rocky Mountain area in relation to relict Laramide uplifts and selected Miocene sedimentary deposits. A–E is line of section for Figure 14. Selected Laramide uplifts are: Du—Defiance uplift; Zu—Zuni uplift; Ku, Kaibab uplift; Mu—Monument uplift; SJu—San Juan uplift; Uu—Uinta uplift. See text for discussion.
|
|
How far to the south, west, and north did the Chuska erg extend? Correlative sedimentary beds of Oligocene age that lack eolian strata are exposed locally along the margins of the Colorado Plateau; these provide maximum paleogeographic limits to the erg margins (Fig. 13). These noneolian Oligocene strata include fluvial deposits of the Mogollon Rim Formation of central Arizona (37–33 Ma; Potochnik and Faulds, 1998) and overlying volcanic and none-olian, volcaniclastic rocks as young as 25.0 Ma (Potochnik, 1989). These rocks were deposited by an alluvial braid-plain sourced from a deep paleocanyon in the Mogollon Highland. Cross-bedded eolian strata may be locally present in the volcaniclastic rocks that overlie the Mogollon Rim Formation (upper part of the Deep Creek section of Potochnik, 1989), but are volumetrically minor. Other noneolian strata include: (1) fluvial and lacustrine deposits of the upper Eocene-lower Oligocene Brian Head Formation and overlying volcanic and volcaniclastic rocks of the Marysvale volcanic field of Utah (Sable and Maldonado, 1997; Eaton et al., 1999; Goldstrand and Eaton, 2001); (2) coarse, fluvial deposits of Oligocene Bishop Conglomerate and correlative fluvial-lacustrine deposits of the Duchesne River Formation (principally the Starr Flat Member) along the south flank of the Uinta Mountains in northeastern Utah (Bryant et al., 1989; Balls et al., 2004); and (3) noneolian, volcaniclastic deposits of the Conejos Formation (ca. 35–30 Ma) in the San Juan volcanic field of southwestern Colorado (Luedke and Burbank, 1963; P.W. Lipman, 2005, written commun.).
We speculate that the northern margin of the erg may have coincided with paleotopography related to laccolith emplacement in the Henry, Abajo, and La Sal mountains (Nelson et al., 1992), and to possible volcanic fields associated with these intrusions (Lipman, 1989). These paleotopographic features may have formed part of the downwind barrier that induced eolian aggradation (see next section). If correct, our reconstruction suggests the Chuska erg covered 140,000 km2. Assuming an average thickness of 200 m, the volume of eolian sand in the erg was 28,000 km3.
|
DEVELOPMENT OF THE CHUSKA ERG
|
|---|
Eolian ergs, or sand seas, record prolonged, areally extensive deposition of sand-sized material largely by eolian dunes. Erg development requires sources of sand that are not stabilized by vegetation or shallow groundwater. Nearly all modern ergs occur in topographic basins or abut high topography (Wilson, 1973; Fryberger and Ahlbrandt, 1979). Eolian deposition occurs where the transportation effectiveness of the wind is diminished by flow deceleration due to atmospheric circulation patterns, vertical expansion of flow into topographic basins, interference of winds by high topography, or by blockage of sand drift by high topography (Wilson, 1973; Fryberger and Ahlbrandt, 1979; Kocurek, 1996). Accumulation of significant thicknesses of eolian sand requires either syndepositional subsidence or rising, shallow water tables (Blakey, 1988; Kocurek and Havholm, 1994; Kocurek, 1996). In this section, we delineate the factors that contributed to the development of the Chuska erg.
During the early to middle Eocene culmination of the Laramide orogeny, fluvial systems flowed onto the southern and central Colorado Plateau from highlands to the south and west, and exited the plateau to the east (Fig. 1; Cather and Johnson, 1986; Potochnik, 1989, 2001a, 2001b; Young, 2001; Cather, 2004). Preserved remnants of the post-Laramide paleogeomorphic surface in southern Colorado (Epis and Chapin, 1975) and on the central Colorado Plateau (Schmidt, 1991; Cather et al., 2003) indicate that relict Laramide paleotopography had been eroded to low relief in these areas by the late middle Eocene. Residual post-Laramide paleotopography in areas of Laramide uplift to the south and west of the Colorado Plateau, however, was significant. Potochnik (2001a, 2001b) demonstrated at least 1000–1400 m of local paleorelief existed on the north flank of the Mogollon Highland in central Arizona during the late Eocene-early Oligocene. In the Grand Canyon region, southwest of the Kaibab uplift, at least 1200 m of late Laramide paleorelief was present (Young, 2001).
In the late middle Eocene to late Eocene, major volcanic centers began to develop along the eastern and southeastern Colorado Plateau margin, including the San Juan, Thirtynine-mile, Ortiz, Sierra Blanca, and Mogollon-Datil fields (Fig. 13). Constructional volcanic topography in these fields appears to have blocked the earlier Laramide drainages that exited the plateau to the east,effectivelytrappingsedimentsontheplateau that prior to volcanism had been transported to the Gulf of Mexico (Fig. 1; Cather, 1992, 2004). Volcaniclastic aprons spread radially from these eruptive centers; some lapped westward onto the Colorado Plateau (e.g., Smith, 2004).
What caused the rapid switch to eolian sedimentation in the earliest Oligocene on the Colorado Plateau? If the beginning of eolian accumulation was regionally synchronous, then the tightest geochronologic brackets on its inception are from the tuff of Luna beneath the eolian section in the western Mogollon-Datil volcanic field (33.51 ± 0.14 Ma; McIntosh et al., 1992) and from a fall-out ash in the basal part of the eolian section in the northern Chuska Mountains (33.54 ± 0.25 Ma; Cather et al. 2003) (Fig. 5). These dates are statistically indistinguishable and are closely compatible with the timing of the Oi1 glaciation/marine isotopic event (e.g., Zachos et al., 2001) and the transition from fluvial to eolian (loessic) sedimentation in the White River Group of eastern Wyoming and Nebraska (Evanoff et al., 1992; Obradovich et al., 1995). Both of these events occurred during magnetic polarity chron C13n (33.761–33.272 Ma, recalibrated from Berggren et al., 1995, using the monitor calibration of Renne et al., 1998 [Fish Canyon Tuff = 28.02 Ma]). The transition from fluvial to eolian (loessic) sedimentation in the White River Group of the Great Plains was accompanied by paleontologic evidence for drying and minor cooling (Evanoff et al., 1992). The Paleogene global transition from greenhouse to icehouse paleoclimates began gradually in the middle Eocene, but was marked by major glaciation of Antarctica and an abrupt, positive oxygen isotopic shift in benthic marine foraminifera (the Oi1 event) near the Eocene-Oligocene boundary (Zachos et al., 2001; Barrett, 2003; DeConto and Pollard, 2003). The near-synchroneity of the onset of eolian sedimentation in the Colorado Plateau-Great Plains region and the Oi1 event suggests that the Oi1 event may have been responsible for aridification of this region.
When the climate became drier at ca. 33.5 Ma, prevailing southwesterly winds began to deflate sediments delivered to the southern and central Colorado Plateau by streams. As sandy eolian environments became more widespread, dunes enveloped fluvial systems on the plateau. This would have necessarily resulted in buried drainages and streams that terminated in ephemeral lakes along the erg margin, as has occurred during the Quaternary in North Africa (e.g., El-Baz, 1998). Eolian sand accumulated over the broad basin of the central and southern Colorado Plateau, eventually attaining local thicknesses in excess of 500 m. Erg sands became intertongued with volcanic and fluvial volcaniclastic rocks as eolianites aggraded and lapped eastward onto the developing volcanic aprons.
In contrast to most erg deposits, eolian aggradation was induced by the effects of constructional volcanic topography downwind of the Chuska erg, not by tectonic subsidence or rising, shallow groundwater. The principal topographic elements east of the Colorado Plateau during the Oligocene were large andesitic to rhyolitic volcanic fields. These fields and their apron deposits formed a relatively continuous highland along the eastern and southeastern margins of the Colorado Plateau (Fig. 13). Thickness data for the proximal portions of volcaniclastic aprons that surrounded Oligocene eruptive centers east of the plateau suggest these aprons stood 0.4–2.3 km above the pre-volcanic surface (e.g., Kautz et al., 1981, Chamberlin and Harris, 1994; Brister and Gries, 1994; Gries et al., 1997; Lipman, 2000). Paleorelief on individual volcanic centers was significantly more than this. Volcanism in most fields adjacent to the Colorado Plateau began before the onset of widespread eolian deposition at ca. 33.5 Ma. Volcanism and associated constructional topographic development continued through most or all of Chuska eolian sedimentation. Volcanic topography caused deposition of eolian sand by diminishing the transport capacity of prevailing winds, perhaps by frictional effects, vertical flow expansion across low-lying areas of the Colorado Plateau, or from the competing effects of catabatic winds related to volcanic highlands. Alternatively, high topography may have simply blocked regional sand drift. Modern examples of eolian dune deposits that have accumulated upwind of topographic barriers include the Great Sand Dunes of Colorado (Johnson, 1967; Andrews, 1981) and the Grand Erg Oriental of North Africa (Fryberger and Ahlbrandt, 1979).
The end of eolian accumulation occurred between 28.2 ± 0.04 Ma and 26.3 ± 0.1 Ma in the western Mogollon-Datil field, and before 25.21 ± 0.16 Ma in the Chuska Mountains. It is possible that the end of eolian sedimentation in the Chuska erg was caused by a change to a warmer and wetter climate, as shown by a marked negative oxygen isotopic shift in benthic marine foraminifera at ca. 27–26 Ma, which has been interpreted to record warming and reduced Antarctic glaciation (Zachos et al., 2001). It is also possible that eolian aggradation on the Colorado Plateau ended due to diminished sedimentary accommodation when constructional volcanism ceased in the adjacent volcanic fields in the late Oligocene. Major volcanism ended in both the San Juan and the Mogollon-Datil volcanic fields ca. 26–25 Ma after voluminous eruptions of mafic lavas (e.g., Chapin et al., 2004).
Because dated volcanic rocks and ashes are intercalated with eolianites in the Colorado Plateau region, we are able to calculate eolian stratal accumulation rates. Assuming that eolian sedimentation began ca. 33.5 Ma and ended ca. 27 Ma, the 535-m maximum preserved thickness of eolian sandstone in the Chuska Mountains corresponds to an overall accumulation rate of 82 m/m.y. The maximum thickness of eolian deposits in the western Mogollon-Datil field is 300 m, indicating a stratal accumulation rate of 46 m/m.y. These overall rates are minima, because they do not account for strata eroded from the top of the eolianite succession.
In the Mogollon-Datil volcanic field, radioisotopic data for specific stratigraphic intervals within the eolianite allow precise calculation of short-term net accumulation rates. Approximately 5 km east-northeast of Mangas, New Mexico, 90 m of eolian sandstone are exposed between the Caballo Blanco Tuff (31.9 Ma) and Vicks Peak Tuff (28.7 Ma) (Chamberlin et al., 1994), yielding a stratal accumulation rate of 28 m/m.y. Eolian sandstone (unit Tss of Ratté, 1989) as much as 60 m thick occurs between the 29.2-Ma Davis Canyon Tuff and the 28.2-Ma Bloodgood Canyon Tuff, indicating a stratal accumulation rate of 60 m/m.y. It is interesting to note that the range of stratal accumulation rates for Oligocene Colorado Plateau eolianites (28–82 m/m.y.) is generally slower than rates of stratal accumulation of Paleocene-Eocene fluvial deposits in the region (65–177 m/m.y.; Cather et al., 1987; Chamberlin and Harris, 1994; Cather, 2004). The slower rates of eolian sedimentation may reflect decreased sediment supply resulting from a drier climate in the Oligocene or from sediment dispersal over a much broader area.
Following cessation of eolian sedimentation in the Chuska erg, eolian sandstones began (or continued) to be deposited in local areas of subsidence adjacent to the Colorado Plateau during the late Oligocene and early Miocene. Examples include the Zia Formation of the Albuquerque Basin (Tedford and Barghoorn, 1999), parts of the Browns Park Formation of the Browns Park Basin and adjacent areas (Honey and Izett, 1988; Buffler, 2003), and eolian dune deposits in the 24.5-Ma Beartrap Canyon caldera of the northeastern Mogollon-Datil volcanic field (C.A. Ferguson, 2006, written commun.). Upper Oligocene-lower Miocene loessic deposits in the Arikaree Group (Hunt, 1990) indicate eolian processes continued to be important in Nebraska and Wyoming after the demise of the Chuska erg.
The widely separated remnants of the Chuska erg attest to the low preservation potential of these strata. Nearly complete erosion of the erg occurred because shortly after deposition of the Chuska Sandstone, a degradational regime ensued that has continued to the present day, interrupted only by the middle to late Miocene aggradation of the Bidahochi Formation (see next section). Due to unrelated factors, Chuska erg deposits are preserved only in three areas. In the Chuska Mountains, cementation of eolianites with opal and chalcedony caused these deposits to be resistant to erosion. Eolian sandstone in the western Mogollon-Datil volcanic field is preserved beneath thick lavas of the Bearwallow Mountain Andesite. In the northwestern Albuquerque Basin, eolianites were preserved by subsidence within the Rio Grande rift.
|
CENOZOIC AGGRADATION AND EXHUMATION ON THE CENTRAL AND SOUTHERN COLORADO PLATEAU
|
|---|
The presence of a thick and extensive Oligocene eolian system on the Colorado Plateau requires significant revision of the Cenozoic geologic history of this area. Although fragmentary, the Cenozoic stratigraphic and paleogeomorphic record of the Colorado Plateau is most complete in its central and southern parts (see summary in Table 5). To evaluate the Cenozoic evolution of the central and southern Colorado Plateau, we have constructed a regional cross section (Fig. 14) that illustrates aspects of Paleocene–Holocene sedimentary aggradation and exhumation relative to several regional chronostratigraphic horizons. The cross section extends from the San Juan volcanic field southwestward across the Laramide San Juan Basin, the Chuska Mountains, and the Bidahochi Formation outcrop to Chevelon Butte, where upper Miocene mafic flows overlie the Upper Triassic strata. There, the line of section bends acutely to the east-southeast through Escudilla Mountain and Escondida Mountain in the northwestern Mogollon-Datil volcanic field. The age of all Cenozoic units encompassed by the cross section is constrained either by fossils or radioisotopic dates.
View this table:
[in this window]
[in a new window]
|
TABLE 5. SUMMARY OF CENOZOIC AGGRADATION AND EXHUMATION ON THE COLORADO PLATEAU, NEW MEXICO, EASTERN ARIZONA, AND SOUTHWESTERN COLORADO
|
|
Two of the key stratigraphic localities used in this study, Escondida Mountain of western New Mexico and Escudilla Mountain of eastern Arizona, lie in the area of structural lowering on the southern Colorado Plateau margin. The southern margin of the Colorado Plateau was structurally lowered relative to the rest of the plateau after the Laramide orogeny (Potochnik, 1989; 2001a, 2001b; Chamberlin and Cather, 1994; Young, 2001). In Arizona, this lowering was the result of Miocene extensional collapse of the southern Basin and Range Province (e.g., Spencer and Reynolds, 1989; Potochnik, 1989). In New Mexico and eastern Arizona, additional components of subsidence may be related to crustal loading by the Mogollon-Datil volcanic field (Chamberlin and Cather, 1994). Along the northwestern edge of the Mogollon-Datil volcanic field, structural lowering was accommodated mostly by gentle southward tilting of Oligocene and older strata, which formed the south limb of a broad regional anticlinorium (Fig. 6). The north limb of the anticlinorium preserves gentle northward dips of Laramide ancestry (e.g., Kelley, 1955; Holm, 2001).
Restoration of the post-Laramide structural lowering is best defined for Escudilla Mountain. By comparing the elevations of stratigraphic markers in the Alpine 1/Federal well on the southwest flank of Escudilla Mountain (Fig. 6) with those in the Belcher 1/State well near the anticlinorial hinge, Witcher et al. (1994) estimated the post-Laramide structural lowering of the Escudilla Mountain area to be 490 m. Based on regional dip data (Chamberlin et al., 1994), we estimate that Escondida Mountain was structurally lowered by 900 m. These estimates are conservative because they only account for removal of post-Laramide southward tilting; no attempt was made to account for the probable former presence of Laramide northward structural tilting in these areas. To maintain an easterly paleoslope (as shown by paleocurrents; Potochnik, 1989) between the projected base of the Mogollon Rim Formation and correlative deposits to the east in the Baca basin, we infer that these deposits have been structurally lowered 430 m.
Post-Laramide Erosion Surface
During waning Laramide deformation in the middle Eocene (Cather, 1990), a low-relief paleogeomorphic surface began to form over parts of the Colorado Plateau. Remnants of this surface are preserved in the San Juan Mountains at 2700-m elevation and the Chuska Mountains at 2350–2440 m. In both areas, the post-Laramide surface is overlain by ca. 35-Ma beds. No analogous erosion surface is present in the Baca basin because volcanism began there prior to the end of Laramide deformation (Cather, 1990). An approximate 35-Ma datum, however, is identifiable within the volcaniclastic deposits of the western Mogollon-Datil volcanic field. This datum (Fig. 14) is at a restored elevation of 3200 m at Escondida Mountain and 2950 m at Escudilla Mountain. The geometry of the post-Laramide surface between Chevelon Butte and the Chuska Mountains is largely inferential. We show it rising to the south and west in accordance with regional Laramide depositional models (e.g., Potochnik, 2001a, 2001b), but slope gradients are poorly defined.
Late Eocene-Early Oligocene Aggradation
By 35 Ma, a regime of widespread aggradation of the central and southern Colorado Plateau was underway. Piedmont deposits prograded southwestward across the inactive San Juan Basin from southwestern Colorado, and are now represented by the Deza Member of the Chuska Sandstone (Cather et al., 2003). The volcaniclastic apron of the Mogollon-Datil volcanic field prograded northward across the axial portion of the Laramide Baca basin (Cather, 1986). The widespread aspect of post-Laramide sedimentation may have resulted from (1) cessation of subsidence in Laramide basins, which could then no longer act as local sediment sumps; and (2) disruption by volcanism of Laramide drainage systems that formerly exited the Colorado Plateau to the east, trapping sediments on the plateau.
Volcanism-induced aggradation may provide an alternative explanation for some sedimentary successions on the southwestern Colorado Plateau that have previously been interpreted as Laramide. For example, the Mogollon Rim Formation of central Arizona is commonly regarded as late Laramide, and has been correlated with the middle Eocene Baca Formation to the east (e.g., Cather and Johnson, 1986). Radioisotopic dates, however, indicate an age range of ca. 37–33 Ma (late Eocene-early Oligocene) for the Mogollon Rim Formation (Potochnik and Faulds, 1998), younger than the Baca Formation but equivalent to the age of the overlying volcaniclastic strata of the Mogollon-Datil volcanic field. Northward progradation of the volcaniclastic apron of the Mogollon-Datil field may have disrupted the paleodrainage of the Baca basin, initiating aggradation of the Mogollon Rim Formation in what had formerly been an area of Laramide erosion and canyon cutting at the western end of the basin.
Similarly, sediments that fill paleocanyons in the Grand Canyon area of the western Colorado Plateau are commonly interpreted to be Laramide (Upper Cretaceous?-Eocene; Elston and Young, 1991; Young, 1987; 2001). This age assignment is based primarily on three lines of evidence. First, these sedimentary units are locally overlain by Miocene volcanic rocks. Second, paleomagnetic measurements indicate the presence of both normal and reverse intervals, which implies these sediments were not deposited during the Cretaceous "long-normal" interval (118–84 Ma). Third, limited fossil evidence has been interpreted to support a late Paleocene-early Eocene age for some of these sediments.
Young (2001) tentatively assigned a late Paleocene to middle Eocene(?) age to the youngest sediments exposed at Long Point, 75 km southwest of the Kaibab uplift in Arizona, based on fossils in interbedded freshwater limestones. These fossils are a charophyte, an ostracod, and viviparid gastropods. Young (2001) listed the charophyte genus Gyrogona as "currently known from early Eocene to middle Oligocene rocks." However, Gyrogona is now known to have a stratigraphic range of Eocene-early Miocene (Feist et al., 2005). Young (2001) tentatively listed the ostracod Bisulcocyridea aravadensis(?) as a late Paleocene to middle Eocene taxon. Indeed, this ostracod, now known as Cypridea (Bisulcocypridea) arvadensis, is known from a handful of North American records of Paleocene-middle Eocene age (Swain, 1999). There are very few documented records, however, of North American nonmarine Paleogene ostracods. For example, in his comprehensive review of published records of nonmarine ostracods from the USA, Swain (1999) listed only two Paleocene records, four Eocene records, and eight Oligocene records. Thus, there is good reason to question whether the stratigraphic range of C. arvadensis has been adequately established.
Young and Hartman (1984, p. 703) referred to probable early Eocene viviparid gastropods from the Long Point limestones. Young (2001, p. 10) listed the gastropod genera present as Viviparus, Pleurolimnacea, Lioplacodes, and Physa and stated that the morphologies of the Long Point specimens overlap those of species assigned
TOP
ABSTRACT
INTRODUCTION
