Oligocene caldera complex and calc-alkaline tuffs and...

Oligocene caldera complex and calc-alkaline tuffs and lavas of the Indian Peak volcanic field, Nevada and Utah

MYRON G. BEST US. Geological Survey and Department of Geology, Brigham Young University, Provo, Utah 84602 ERIC H. CHRISTIANSEN Department of Geology, Brigham Young University, Provo, Utah 84602 RICHARD H. BLANK, JR. US. Geological Survey, Denver, Colorado 80225

ABSTRACT

The Indian Peak volcanic field is representative of the more than 50,000 km3 of ash- flow tuff and tens of calderas in the Great Basin that formed during the Oligocene-early Miocene "ignimbrite flareup" in southwestern North America. The field formed about 32 to 27 Ma in the southeastern Great Basin and consists of the centrally positioned I n d i Peak caldera complex and a surrounding blanket of related ash-flow sheets distributed over an area of about 55,000 km2. The field has a volume on the order of 10,000 km3. A cluster of two obscure source areas and four calderas comprise the -80 x 120 km caldera complex. Only minor volumes of rhyolite and two pyroxene andesite lavas were extruded episodically throughout the lifetime of the magma system that formed the field, chiefly during its youth and old age.

Six ash-flow sequences alternate between rhyolite and dacite in a volume ratio of about 1:8, and a culminating seventh is trachytic. The first, fourth, and sixth tuff units are of rhyolite that contains sparse to modest amounts of phenocrysts, chiefly plagioclase and biotite, and abundant lithic and pumice lapilli; these deposits are confined within the caldera complex and form multiple and compound cooling units that are normally zoned with respect to bulk chemical composition and crystal type, content, and size. The sec- ond, third, and fifth tuff sequences are of crystal-rich dacite that forms extensive simple cooling-unit outflow sheets and partial caldera fillings of compound cooling units. Each dacite unit contains similar amounts of plagioclase, biotite, hornblende, quartz, two pyroxenes, and Fe-Ti oxides; trace amounts of sanidiine and titanite also occur in the youngest. Cognate inclusions in the dacites show only slight intra- and inter-unit differences in bulk chemical composition. The seventh

eruptive sequence consists of several cooling units of trachydacite tuff containing small to modest amounts of plagioclase and two pyroxenes.

These dominantly high-K calc-alkaline rocks are a record of the biih, maturation, and death of a large, open, continental magma system that was probably initiated and sustained by influx of mafic magma derived from a southward-migrating locus of magma production in the mantle. The small volumes of chemically diverse andesitic rocks were derived from separately evolving magma bodies but are modified representa- tives of the mantle power supply. Recurrent production of very large batches (some greater than 3,000 km3) of quite uniform dacite magmas appears to have required combi- nation of andesite magma and crustal silicic material in vigorously convecting chambers. Compositional data indicate that rhyolites are polygenetic. As the main locus of mantle magma production sh ied southward, trachydacite magma could have been produced by fractionation of andesitic magma withiin the crust.

INTRODUCTION

We document a large-volume, cyclic eruptive sequence of rhyolite, dacite, and trachydacite ash-flow tuffs and coeval andesite and rhyolite l ava comprising the Indian Peak volcanic field astride the southern Utah-Nevada state line. Tuffs were derived from a centrally positioned magma locus marked by the Indian Peak caldera complex, a cluster of four known calderas and two inferred source areas. The dacite tuffs are representative of the Monotony compositional type of tuff which dominates the late Oligocene of the Great Basin (Best and others, 1989a; compare the "monotonous interme- diates" of Hildreth, 1981). The origin of similar crystal-rich dacite tuff in the San Juan field is the

subject of recent controversy (Whitney and Stormer, 1985; Johnson and Rutherford, 1989; Grunder and Boden, 1987).

Compared to contemporaneous volcanic fields around the Colorado Plateaus to the east (Fig. l), the -10,000 km3 of ash-flow deposits in the Indian Peak volcanic field is an order of magnitude greater than in the Marysvale field (Steven and others, 1984) but similar to that of the San Juan (Steven and Lipman, 1976) and Mogollon-Datil fields (RattC and others, 1984). In these three fields, however, individual ash- flow eruptions were on the average smaller, and more calderas formed than in the Indian Peak. The Indian Peak field also differs from the other three in its overwhelming dominance of silicic ash-flows over intermediate lava flows, a characteristic aspect of Oligocene-early Miocene rocks of the Great Basin (Best and others, 1989a).

This paper summarizes the eruptive history of the Indian Peak volcanic field and the basic compositional characteristics of the rocks and concludes with a provisional history of the magma system that provides a basis for continu- ing more detailed analytical studies of its ther- mochemical evolution.

REGIONAL GEOLOGIC SETTING

Tertiary volcanic rocks are virtually the only magmatic rocks of Phanerozoic age in the southeastern Great Basin. They were deposited on an erosional surface of modest relief carved into a thick, upper Proterozoic through lower Mesozoic miogeoclinal sedimentary sequence. During the Cretaceous Sevier orogeny, this sequence was folded and thrust-faulted and subse- quently locally overlain by lower Tertiary sedimentary deposits (Stewart, 1980). A grow- ing body of data documents episodes of locally extreme crustal extension (as much as 300%) in the middle Cenozoic era (for example, Wust, 1986) before, during, and after volcanism and generally before the classic high-angle Basin and

Geological Society of America Bulletin, v. 101, p. 1076- 1090, 8 figs., 3 tables, August 1989.

INDIAN PEAK VOLCANIC FIELD, NEVADA AND UTAH

\ I ANDESlTlC LAVA PLATEAUS \ a AND DEBRIS FLOWS \ 1

, FELSIC INTRUSIVE ROCKS

L CALDERA MILES 400 I 0 KILOMETERS

Figure 1. Generalized distribution of Oligocene and lower Miocene (mostly 34 to 17 m.y. old) magmatic rocks and known calderas in the Great Basin of Nevada and western Utah and in the . . . . Marysvale, San Juan, and Mogollon-Datil volcanic fields. The Mogollon-Datil field includes considerable intermediate composition lava flows which are not separately distinguished. Data from Lipman (1984), Ratti! and others (1984), Sargent and Roggensack (1984), Steven and others (1984), and Stewart and Carlson (1976). Dotted line is approximate margin of Colorado Plateaus. - - -

Range normal faulting and extension in the late Cenozoic era. Many recent interpretations of the total amount of this more or less east-west extension suggest that it varied greatly from place to place through time (for example, Coney and Harms, 1984; Wells and Heller, 1988; Taylor, in press), precluding a reliable single-value estimate of the stretching of Great Basin tuff sheets. We use a figure of 50% east-west crustal extension since deposition in restoring the distribution of the different sheets and in calculating their origi- nal volumes.

The Cenozoic volcanic history of the central western United States is dominated by a broad southward sweep of essentially calc-alkaline igneous activity (Cross and Pilger, 1978) while oceanic lithosphere was subducting beneath the western margin of North America. Between about 30 and 25 Ma, this transgressive activity decelerated or even stagnated in the southern Great Basin (Best and others, 1989a) where the Indian Peak volcanic field developed. This

broad zone of voluminous late Oligocene-early Miocene volcanic rocks extends across Nevada and Utah into southwestern Colorado (Fig. 1). Within this zone, chiefly calc-alkaline, highly potassic intermediate to silicic volcanism was gradually supplanted by bimodal basalt-rhyolite volcanism in the late Cenozoic era. Within the eastern Indian Peak volcanic field, early, middle, and late Miocene episodes of generally high- silica rhyolitic activity were accompanied by extrusion of mafic lavas, but not until the last episode had their content of K 2 0 and Si02 de- creased to the level of true basalt (Best and others, 1980, 1987~).

GEOPHYSICAL EXPRESSION O F THE CALDERA COMPLEX

Although bearing the overprint of basin and range faulting, gravity data delineate the -80 x 120 km Indian Peak caldera complex (Fig. 2). None of the Oligocene calderas has any topo-

graphic expression in the present terrain which is dominated by northerly trending basins and ranges produced by subsequent block faulting; nearly half of the caldera complex now lies bur- ied beneath alluvium-filled basins. Most impor- tantly, gravity gradients mark the southwestern and southern margin of the complex which is apparently filled with thick, low-density tuff or underlain by low-density intrusive rocks. With- out the gravity data, the southern margin can be only indirectly located, even in the mountain ranges, because of an extensive cover of younger volcanic rocks.

Aeromagnetic data (Zietz and others, 1976, 1978) also disclose the general location of the Indian Peak caldera complex.

Geologic mapping corroborates the compound nature of the geophysically expressed caldera complex. Topographic margins of four nested calderas are indicated by striking discon- tinuities in thickness of tuff deposits and by coarse breccia of older rocks shed off caldera

1078 BEST AND OTHERS

walls. Short segments of caldera-bounding faults are locally exposed or indirectly expressed by small dacite lava domes. The sources of two additional ash-flow sheets are approximately located by the distribution of the sheets and by clast size.

ERUPTIVE HISTORY

The Indian Peak volcanic field was dominated by eruption of voluminous ash flows (Figs. 3,4, and 5); lavas are much less extensive, and deposits of pyroclastic surge and fall are rare. Rock analyses (Tables 1,2, and 3) indicate that the volcanic field is typical of the Great Basin as a whole (Best and others, 1989a); the rocks have relatively low Fe/Mg ratios and

generally high K 2 0 concentrations that qualify them as a high-K calc-alkaline suite (Ewart, 1979).

Voluminous eruption of lavas did not precede nor accompany ash-flow eruptions from the In- dian Peak complex, in contrast to other nearly contemporaneous volcanic fields in the southwestern United States (Marysvale, Utah- Steven and others, 1984; Mogollon-Datil, New Mexico-RattC and others, 1984; and San Juan, Colorado-Lipman, 1975; Fig. I). Had large piles of lava existed, it is unlikely that all of the edifices as well as peripheral alluvial deposits would have been completely engulfed. Exten- sive lavas do not occur in resurgent parts of the calderas, which locally expose Paleozoic rock on the caldera floor. Only small volumes of

0 20 miles 40 60 I - - - 4

0 20 km 40 60

Figure 2. Bouguer gravity in and around the Indian Peak caldera complex. Southwestern comer of the map area and area north of about 38" show horst-graben structure of the Basin and Range province (compare with topographic features in Fig. 4). Thick accumulation of relatively low density tuff filing two and locally three superposed calderas produces gravity lows along the major west-northwest axis of the elliptical caldera complex. Identified calderas based on unpublished and published geologic mapping of M. G. Best and associates (see References Cited) are, from oldest to youngest: Pme Valley, PV; Indian Peak, IP; White Rock, WR; Mount Wilson, MW. Areas from which the Cottonwood Wash (CW) and Isom (IS) tuffs were probably derived are denoted by diagonal ruling (compare Figs. 5A and 5D). Assumed crustal density is 2.67 g/cm3; contour interval is 10 mgal, with progressively more negative

35-34-m.y.-old rhyolitic tuffs and lavas were emplaced in and near the volcanic complex prior to its development (Best and others, 1987%; Loucks and others, 1989).

Volumes of ash-flow tuff deposits in Figure 3 were calculated using restored areas based on 50% east-west crustal extension and the method of Ekren and others (1984). The lack of a tight control on the amount of extension is a possible major source of error in the listed areas and volumes. For example, the estimated volume of the outflow member of the Wah Wah Springs Formation is about 1,500 km3 if an east-west extension of loo%, which we consider excessive, is used. Several additional factors also contribute to the uncertainty of areas and volumes, as follows. (1) Edge of outflow sheet may have been eroded before deposition of younger sheet. (2) Areas of older calderas engulfed by younger are uncertain. (3) The thickness of intracaldera tuff now obscured by resurgence, basin-and- range faulting, and younger deposits in known calderas and concealed sources is uncertain. Vol- ume estimates of intracaldera dacite tuff in Figure 3 are probably conservative; volumes of Cottonwood Wash and Isom tuffs do not include possible thick accumulations within their sources. (4) The existence of predeposition to- pography created variably thick deposits, especially of older tuff sheets. (5) A final factor is the porosity of minor weakly and rarely nonwelded parts of some outflow sheets that has not been compensated to a dense-rock equivalent.

Escalante Desert Formation and the P i e Valley Caldera

Initial eruptions about 32 Ma from the Indian Peak caldera complex produced several hundred cubic kilometers of a bimodal assemblage of rhyolite tuff and minor rhyolite and two- pyroxene andesite lava of the Escalante Desert Formation (Fig. 3; Best and Grant, 1987).

The crystal-poor Marsden Tuff Member is almost everywhere propylitically altered and is generally characterized by conspicuous xenoliths of sedimentary rock and rhyolite; variations in their size and concentration indirectly indicate the location of an obscure source (the Pine Val- ley caldera) largely engulfed in younger calderas (Best and Grant, 1987). Phenocrysts consist of plagioclase and minor quartz, sanidine, biotite, Fe-Ti oxides, apatite, and zircon (Fig. 6); these increase in abundance upward.

The younger Lamerdorf Tuff Member contrasts with the Marsden in its greater areal extent (Best and Grant, 1987), generally thinner, more densely welded cooling units that have local basal vitrophyres, more phenocrysts of larger size, and conspicuous xenoliths of rhyolite. The areas below -200 mgal more strongly stippled.

INDIAN PEAK VOLCANIC FIELD, NEVADA AND UTAH

Figure 3. Stratigraphic relations and ages of rocks in the Indian Peak volcanic field (Best and Grant, 1987, average K-Ar age of 26 m.y. of the Isom For- mation cited in this reference appears to be at least 1 m.y. too young based upon 4 0 ~ r / 3 9 ~ r determinations of Alan Deino, 1988, written commun.). AU units except Isom belong to Needles Range Group. Vertical scale is not linear and shows only stratigraphic position. Areas in km2 and volumes in km3. Lower of two values for tuff member of Lund Formation is volume outside White Rock caldera; upper value is for tuff inside caldera based on a conservative estimate for thickness of 1 km; locally it may be as much as 3 km.

Lamerdorf contains minor amounts of clinopyroxene, hornblende, Fe-Ti oxides, apatite, and zircon, in addition to more abundant plagioclase and biotite.

Rhyolite and andesite lava flows are more widespread in the Escalante Desert Formation than in the younger formations within the caldera complex and occur chiefly in and near its north and east sectors.

Andesite lavas are generally nonvesicular, two-pyroxene plagioclase rocks, but phenocrysts of biotite and hornblende also occur in lavas in the southern Fortification Range. Andesite lavas include both high (greater than about 5%) and low MgO varieties (Figs. 7 and 8); the magne- sian andesites also have higher K 2 0 and lower Al2O3 concentrations.

Chemical analyses of the rhyolites of the Es- calante Desert Formation show that the Lamer- dorf tuff and most of the lavas are low-Si02 rhyolite and even dacite (67% to 72% SiOz; Table 2 and Fig. 7). The Lamerdorf tuff displays slight vertical chemical zonation with Si, K, and Rb depleted and Al, Fe, Ca, Sr, Zr, and Ba enriched in the upper part of the tuff. Most samples of the rhyolitic lavas and Lamerdorf tuff are enriched in K, Ba, Ti, and Zr and depleted in Ca compared to younger rhyolites (Fig. 8).

Cottonwood Wash Tuff

The next eruptive event at 30.6 Ma formed the Cottonwood Wash Tuff, generally a simple

cooling unit of crystal-rich dacite ash-flow tuff. Thickest sections (as much as 300 m, Fig. 5A) are commonly marked by gray, near-basal vitrophyres. The distribution and thickness of the tuff, its vitrophyre, and local lithic clasts suggest a source between the northern Needle and Forti- fication Ranges and south of the Snake Range. These ranges themselves disclose no direct evidence of a nearby fault-bounded depression. Nonetheless, gravity data (Fig. 2) permit a bur- ied source for the - 1,500 km3 of tuff.

The Cottonwood Wash Tuff has a phenocryst assemblage dominated by plagioclase and un- commonly large books of biotite (up to 8 mm across); embayed bipyramidal quartz; hornblende (small grains inconspicuous in hand sample), two pyroxenes, Fe-Ti oxides, apatite, and zircon make up the rest (Fig. 6). Like other dacites in the Indian Peak field, the Cottonwood Wash has relatively low Zr concentrations for its Si02 content (Fig. 8).

Wah Wah Springs Formation and Indian Peak Caldera

Emplaced at about 29.5 Ma, the tuff of the Wah Wah Springs Formation comprises a widespread simple cooling unit of crystal-rich dacite that ranges up to 520 m thick outside of its resurgent Indian Peak caldera source (Best and Grant, 1987; Figs. 2, 4, and 5B). The intracaldera member of the formation consists of breccia, intrusive rocks, and a thick compound

cooling unit of dacite tuff similar to the outflow tuff member but containing conspicuous lithic fragments and variable but commonly slightly more abundant phenocrysts of quartz (Best and Grant, 1987; Best and others, 1987a). Together, the volume of the intra- and extra-caldera tuff is on the order of 3,000 km3.

Less quartz, more hornblende, and smaller phenocryst size (especially biotite) distinguish the Wah Wah Springs from the otherwise similar Cottonwood Wash Tuff (Fig. 6). There are only minor differences in the chemical compositions of bulk samples of these two high-K calc- alkaline dacites (Table 3; Figs. 7 and 8).

Little andesite lava was erupted during the middle history of the volcanic field when voluminous eruptions of dacite occurred. Only small outcrops of andesite of possible Wah Wah Springs age have been found within the caldera complex (Table 1). Sixty kilometers north of the caldera complex, an olivine-bearing, high-MgO basaltic andesite lies between the Cottonwood Wash and Wah Wah Springs tuffs. This lava shows compositional similarity to those in the Escalante Desert Formation (Table I).

Ryan Spring Formation

The Ryan Spring Formation consists mostly of several hundred cubic kilometers of variably welded rhyolite tuff that is virtually identical in hand sample to the rhyolite tuffs of the Escalante Desert Formation. The basal or Greens Canyon


Tuff Member of the Ryan Spring Formation commonly contains clasts of Wah Wah Springs tuff instead of sedimentary xenoliths as is characteristic of the otherwise similar but older Marsden tuff. In the moat of the resurgent In- dian Peak caldera in the northern White Rock Mountains (Best and others, 1989b), the Greens Canyon Tuff is a wmpound cooling unit as much as 1.4 km thick that consists of five wol- ing units with local vitrophyres near their bases. Directly overlying the Greens Canyon is the upper, or Mackleprang Tuff Member of the Ryan Spring Formation that is more strongly porphyritic and in most places cannot be distinguished in outcrop from the Lamerdorf Tuff Member of the Escalante Desert Formation; in the field, these "twins" can be distinguished only by reference to stratigraphic position.

Tuff of the Ryan Spring Formation is almost wholly confined within the Indian Peak caldera (Best and Grant, 1987; their Figs. 4 and 5) which probably also wntains its source. In the Atlanta, Nevada, open-pit gold mine, thick, altered tuff is banked against silicified Paleozoic carbonate rocks (Willis and others, 1987). Lo- cally derived slabs of brecciated Paleozoic rocks appear to lie on the tuff, suggesting recurrent collapse of the Indian Peak caldera as tuff of the Ryan Spring Formation was erupted.

The similarities in age, distribution, and mineralogy suggest that the Greens Canyon and Mackleprang tuffs are, respectively, the lower and upper parts of a single eruption sequence that is wmpositionally normally zoned. The Greens Canyon is slightly more evolved (lower Ti, Al, Fe, Mg, and Zr and higher Si) than the

Mackleprang (Table 2) and, like many normally zoned ash-flow deposits, it contains fewer and smaller phenocrysts (Fig. 6). The Mackleprang Tuff lacks the sparse quartz and sanidine phenocrysts found in the lower unit. Compared to the similarly zoned Lamerdorf Tuff and the rhyolite lavas of the Escalante Desert Formation, the Ryan Spring tuffs have slightly lower concentrations of Ti, Fe, Mg, Y, Zr, and Ba (Fig. 8). Like the Lamerdorf Tuff, however, the Ryan Spring tuffs are low-SOz, high-K, calc-alkaline rhyolites (Fig. 7).

Minor volumes of andesite that occur beneath the Mackleprang tuff in the southeastern segment of the Indian Peak caldera have been as- signed to the Ryan Springs Formation. These two-pyroxene plagioclase andesites (Table 1) locally contain hornblende.

Figure 4. Present maximum extent of major tuff sheets of the Indian Peak volcanic field around the caldera complex (heavy line, dashed where less certain around obscure source areas). Topographic contours at 5,000,7,500, and 10,000 fi. Compare Figures 2 and 5.

Formation Escalante Deren Wah Wah Springs Ryan Spring Lund lsom Estimated -- accuracy

Sample 1 2 3 4 5 6 7 8 9 10 I1 12 ( i 2 sigma)

Field number

LAM-9 MIL-l TET-9 TET-9 BRNJ BUCK-8 MIN-8 STM-8 STM-8 STM-8 STM-8 GLE-6 -71-2 - 13-2 -7748 -140-2 -23-1 -169-3 -1694 -128-1 -130-1 -961

Si02 Ti02

A'2°3 Fe203 MnO MgO CaO Na20

K20 '2'5 TOTAL

I. Glassy matrix; lat. 38'18'48*, long. I 13'30'00-. 2. Holwslalline: uwer flow in andsite vile immediatelv belnw Cottonwood Wash TuE lat. 38'17'58-. lone. 113"13'3'. 3. ~li~htl;alteral, nbw ahnr Waodcn T;R Member, la; 38°1).16.. long 1 13°34'16- 4. H o l ~ r r a l l ~ n c , flow ah,ur Warden Tuff Mrmbcr: la1 38°10'58'. long 1 13°30'31'. 5. Basaltif andesile lava flow comprised of phenccrysts of olivine in a mat& ofplagiocl~, awte, glass, and magnetite. betwecn oufflow tuff member of Wah Wah Springs Formation and Cottonwood Wash Tuff; lar 38°50'20". long. 113'28.25'. 6. Altered matrix. fr& phencctysu; lat. 38'16'20". long. 113'51'42". 7. Holocmralline flow wntainine oaniallv chloritmd ovroxene ohenccmts on townraohic rim of Indian Peak caldera: lar. 38O22'4". lone. 113"56'6". . . . . . . . . - . S Glassv matnk. abmdant hornhlrnde and mmor blot~tc phenwr).cn nllh plagicclw and two pyrorcmes, lal. 38'2'41-, long. 1 13O48'52'. 9. Holccrystallinc, sltghtl, altered hyponthenr @hmccr)~s: immd~ately ahove luff; a hsx of a n h u e pdc. IJI 3812'?4', long. 113'49'19- . .

10. Glassy matrix, at top of andesite pile; lat. 3Rb1'58", long. 113~50'00'. I I. Strong porphyritic, glmy matrix; lat. 38'239-, long. 1 13°50'10-. 12. Holoc~ralline, sparse hornblende phenccrysu; below tulis; lat. 38"lCSO'. long. 114"6'2-. Oxide values have been recalculated to 100% on a wlatile-free basis. Analytical total for 10 oxides in parentheses.

TABLE 2. CHEMICAL COMPOSITION OF RHYOLITE LAVA FLOWS AND TUFFS

Stratigraphic Escalante Desert Formation Ryan Spring Ripgut unit Formation Formation lsom Formation

Rhyoltte lava flow member Lamerdof Tuff Member Lower Upper Lower Upper Greens Mackleprang

Canyon Lower Upper Lower Upper Sample I 2 3 4 5 6 7 8 9 10 I1 12 13 14 15 16

Field MIN-8 MSP-9 TET-9 TET-9 HFW- HFW- MLLR-6 MIN-8 ATL-l MTWlLS STM-8 MLLR-6 MLLR-6 MLLR-9 number 40-10 -37-1 43-3 -75-1 I A 2A I -61-2 -68-1 50 -169-5 -17-1 -17-2 -164-3

Si02 T,02

A1203 Fe203 MnO

ME0 c a o Na20

K20 '2'5 TOTAL

I. Felsitis lat. 38°17'42", long. 1 1395'00'. 2. Fclsitiq lat. 38'3'18". long. 113'31'40". 3. Felsitif; lat. 38°13'12", long. 1 13°34'17". 4. FeUtif; lat. 38°13'18", long. 11391'49". 5. Average of 3 analyses from Campbell (1978). 6. Basal vitmphyre; lat. 3890'32", long. 113'50'57". 7. Average of 4 analyses from Campbell (1978). 8. Basal vitmphyre: lat. 38'30'32.. long. 113'50'57". 9. Basal vitmphyre; la1 38'17'43.. long. 1 14O9'45".

10. Devitrilied base of unit; lat. 38'20'5". long. 113"55'30'. I I. Basal vitrophyre; lat. 38'257'. long. 114'2'2". 12. Felisitc; 3X014'6", long. 114'22'40". 13. Felsitic; tat. 38'1'43". long. 113"48'57". 14. Felsitic; poikilitic biotite in matrix: oldest of 3 cooling units; lat. 38'16'24". long. 114'1 1'25' 15. Felritic; youngest of 3 cool~ng units; lat 38°16'19', long. 114°1 1'14'. 16. Basal viaophyre; lal. 38'15'18", long. 114°12'55".

BEST AND OTHERS

0 K ILOMETERS 6 0

Figure 5. Present extent and thickness (100-m contours) of major tuff sheets around source calderas. Zero (0) indicates unit is absent between older and younger units. Di- agonally ruled areas in 5A and 5D are probable source areas. Dotted l i e is outer perimeter of Indian Peak and White Rock calderas (compare with Fig. 2). Heavy line is topographic margin of caldera related to particular tuff. In 5B, inner ring fault of the Indian Peak caldera is orna- mented by ball and bar. In 5C, stars denote flow-layered dacite bodies of Lund mineralogy. Thickness of Isom tuff south of its source partly from unpublished mapping of R. E. Anderson.

I -

/ ? - - I / ?30\,

\ B.

W A H W A H SPRINGS

1 1 2 O + 39O

KILOMETERS 6 0

Lund Formation and the White Rock Caldera Lund tuff appears to be similar to that of the Wah Wah Springs Formation (Fig. 3).

Eruptions at 27.9 Ma produced another The perimeter of the White Rock caldera is voluminous crystal-rich dacite tuff that consti- indirectly marked by gravity gradients on the tutes most of the Lund Formation and created south and west (Fig. 2) and by geologic relations the resurgent White Rock caldera that engulfed elsewhere (Best and Grant, 1987). The topo- the southwestern part of the older Indian Peak graphic margin is well exposed in the northern caldera (Figs. 2 and 5C). The volume of the White Rock Mountains (Best and others,

1989b) where massive Lund and younger R i p gut and Isom tuffs that fill the caldera thicken southward from a few tens of meters to as much as 2 km. In the northern Wilson Creek Range (Willis and others, 1987), the topographic margin is marked by the southward thickening tuff member of the Ripgut Formation. Beneath the Ripgut, breccia of Wah Wah Springs and Ryan

INDIAN PEAK VOLCANIC FIELD, NEVADA AND UTAH 1083

1 so 115' + LUND +

.O

I .o

0 MILES 5 0 .o

0 KILOMETERS 6 0

.o .o

Figure 5. (Continued).

I '0 115' 113O 112 + <I+ + + 39O

ISOM

D. .-J

I I '5 - 0

' 0 .o /

100 \ ?8

745 5 .150+ .420

?18

a 0 MILES 5 0

100 0 KILOMETERS 6 0

.o + + + 37O

Spring tuffs, Paleozoic carbonate rocks, and, lo- tuff and breccia in the Fairview Range (20 km placed along the caldera-bounding fault marking cally, granitic rock occur as wedges intercalated due west of Mount Wilson; Fig. 4). the structural margin of the caldera. within intracaldera Lund tuff. If unfaulted, the Five small masses of porphyritic and locally Southward from Atlanta, Nevada, the young- intracaldera Lund tuff and breccia could be as flow-layered dacite similar to the Lund tuff er Ripgut tuff thickens and then thins and much as 3 km thick in the northern Wilson occur inside the northern topographic margin of pinches out north of Mount Wilson (Willis and Creek Range. Reconnaissance work has re- the caldera (Fig. 5C). These are interpreted to be others, 1987). This lens-shaped mass of tuff de- vealed a similar intracaldera sequence of Lund shallow intrusions and lava-flow remnants em- fines a moat between the topographic wall of the

BEST AND OTHERS

TABLE 3. CHEMICAL C O M P O S ~ O N OF DAClTE TUFFS

Stratigraphic Cottonwood Wash Tuff Wah Wah Springs Formation Lund unit Formation

OuUlow tuff member lntracaldera member

Sample I 2 3 4 5 6 7 8 9 LO LI 12 13 14

Field HFW-8 HFW-8- ERN-2 ERN-2 HFW-8 HFW-8- BRN-I ERN-I HAM-9- MLLR-6- MLLR6 GLE-6- MINS-8- MLLR-9- number* 153-IBV 153-ICD M P 153-3BV 153-3CD M P 129-IP 63-IP 64-2X 98-IX 61-4BD 164-1X

Si02 Ti02

A1203 Fez03 MnO MgO CaO Na20

K20 '2'5 TOTAL

*V, vitrophyre at base of ash-flow sheet; D, devivified densely welded just above basal vitmphyre; P, pumice lapilli or block in tuff; X nonvesicular vitrophyre block in tuff; M, tuff matrix enclosing pumice lumps. I. Basal vitrophyre; lat. 3S033'25-, long. 1 13°51'20" 2. Devitrified densely welded tuff immediately above basal vitmphyn; laL 3E033'25', long. 1 13°5L'2W . . 3. Matnx of IuRaround pumace blocks near baw of weakly welded unit la!. 38°4YlW, long. 113°28'30' 4. Pum~ce blwk from ueaklr welded base of una la!. 38'49'10'. lone. 11398'W . - 5. Basal vitmphyre; laL 38"34'34', long. 1 13°51'26- 6. ~evitrifieddknsel~ welded tuff immediately above basal vitmphyre; lat. 3S034.34", long. 1 13°51'20" 7. Matrix of tuff around ~umice laoilli 4-6 m above base of weallv welded unit lat. 3S048'W. lone. L13"27'30" . - 8. Pumice lapilli 4-6 m above bsse of weakly welded unit lat. 38'48'~, long. I1397'W 9. Pumice block from weakly welded top of unit lat. 38°24'12', long. I l4W35-

10. Pumice block from weakly welded top of uniS lat. 38°18'51", long. 114"8'48' I I. Nonvesicular vitrophyre block within a few meten of top of unit lat. 38"18'37', long. 114097" 12 Nonva~cular vtlrophyrc block wttbtn a feu meten of top of un4 lat 38°17'57'. long 114°6'45' 13 Dcvtlnfied dcnul) welded tuff tmmed~atelv above basal vllru~hvre. kL 38'20'5'. lone 1 13°55'W . . 14. Nonvericular viwophyre block in tuff in probable vent mass; lal. 38'15'16', long. 114'12'24"

White Rock caldera and its resurgent wre to the south.

east of the Nevada-Utah state line in NW% sec. 36, T. 29 S., R. 20 W. in the northeastern wrner of the White Rock Peak 7% minute quadrangle. The type area for the Ripgut Formation is the western slope of Mount Wilson in sections 25, 26, 35, and 36, T. 5 N., R. 67 E. along the

East of Mount Wilson, on the northeast side of the Mount Wilson caldera, a clastic member

The most notable wmpositional difference overlies the tuff member of the Ripgut Forma- tion. These accumulations of bedded tuff, silt- stone, sandstone, and local conglomerate aggre- gating no more than a few tens of meters in

between the Lund and the older dacite ash-flow deposits in the Indian Peak volcanic field is the small amount of sanidine and titanite. Quartz is more abundant and clinopyroxene and ortho- southern margin of the Schoolmarm Basin 7% thickness represent another post-rhyolite erup- pyroxene are sparse; they are found only as anhedral wres within hornblende phenocrysts. The major-element composition of the Lund tuff

minute quadrangle, Nevada (Willis and others, 1987). Here, along the northern wall of the Mount Wilson caldera that was the source of the

tive hiatus. Only the northern segment of the Mount Wil-

son caldera is clearly defined around its name- sake in the northern Wilson Creek Range (Fig. 2; Willis and others, 1987). Mount Wilson itself is underlain by as much as 2 km of the tuff member of the Ripgut Formation. Southward in the Wilson Creek Range, as much as several

lies within the range observed for the other da- tuff member of the formation, southward- cites from the magma system; however, there are small differences in Rb, Sr, Zr, and Ba (Table 3; Fig. 8).

thinning wedges of the breccia member are intercalated with the tuff member. These two members, which comprise the intracaldera sequence, overlie the tuff member of the Lund Ripgut Formation and the Mount

Wilson Caldera Formation and in turn are overlain by the Blawn Formation (Best and others, 1987~). The breccia member is composed of clasts of Lund tuff shed

hundred meters of rhyolite tufk and lava flows of the Blawn Formation (Best and others, 1987c) deposited about 23 to 18 Ma wnceal the A third sequence of rhyolite ash flows erupted

from within the caldera wmplex after deposition of the Lund tuff and possibly well before emplacement of the Isom Formation about 27 Ma. These eruptions formed several hundred cubic kilometers of tuff and created the small Mount Wilson caldera. This tuff and associated landslide breccias and epiclastic deposits within the source caldera comprise the Ripgut Forma- tion, a new formal unit defined here. The name is taken from Ripgut Springs located about 1 km

off the topographic wall of the Mount Wilson southern margin of the Mount Wilson caldera. As defined by Best and Grant (1987), the

youngest unit in the Needles Range Group is the caldera and has either a matrix of Ripgut tuff or comminuted Lund tuff. An easily accessible, well-exposed reference section for the tuff Lund Formation. Since that report was written, member of the Ripgut Formation is in the Atlanta 7% minute quadrangle north of White Rock-Bailey Spring to the top of hill 8046

however, the Ripgut Formation has been found to be an integral part of the Indian Peak volcanic field and is therefore now included as the

which is capped by Isom tuff. This section is in an unsurveyed area at 38'25'10"N and 114O22'5"W.

youngest formation in the Needles Range Group.

The tuff member of the Ripgut Formation is


rhyolite and petrographically similar to the older Marsden and Greens Canyon tuffs (Fig. 6). As in other rhyolite tuffs in the complex, volcanic xenoliths are locally abundant; in the Ripgut, these clasts are of Wah Wah Springs and Lund tuffs. As many as three ash flows comprise the compound cooling unit in which intensity of welding and compaction range from dense, black or dark brown vitrophyre as much as 40 m thick upward into a nonwelded, glassy lapilli and ash mixture.

The sequence of Ripgut cooling units is wmpositionally zoned (Table 2 and Figs. 6 and 8). The basal unit has only a few percent small phenocrysts of plagioclase, biotite, quartz, sanidine, and Fe-Ti oxides. This unit is a high-Si02 rhyolite with low concentrations of CaO (0.7 wt%), MgO (0. I%), Ti02, Fe203, P2O5, Sr, and Zr (Table 2 and Fig. 8). The upper unit on Mount Wilson has about 5% phenocrysts of pla-

gioclase, biotite, hornblende, quartz, and trace amounts of titanite and Fe-Ti oxides. It is a low- silica rhyolite like the Ryan Spring tuffs.

Isom Formation

Trachytic ash flows erupted about 27 Ma formed one to three tuff cooling units of the Isom Formation (Anderson and others, 1975) in and near the caldera complex. The distribution and thickness of the formation (Fig. 5D) and size of pumice clasts point to a source north of Modena as suggested by Rowley and others (1979). Widespread Miocene volcanic rocks (Best, 1987) and alluvium, however, conceal the source area of the more than 1,300 km3 of pyroclastic material. Based on this overlapping source area, a time lapse on the order of only 1 m.y. since eruption of the Lund tuff and accord- ingly less for the Ripgut tuff, and their high-K

calc-alkaline compositions, it appears that the Isom magmas were derived from the Indian Peak system.

Cooling units are everywhere densely welded and commonly have black basal vitrophyres a meter or two thick. Vugs partly filled with vapor-phase minerals are typical; both vugs and pumice lumps are commonly flattened and elongated, presumably due to secondary flowage after deposition. In some thick sections within the Indian Peak caldera complex, flow units of Isom mineralogy that lack eutaxitic texture may be lava-like, ash-flow tufk formed by secondary flowage after deposition on an uneven surface (compare Ekren and others, 1984). In some places within the southern part of the White Rock caldera, Isom tuffs have intensely com- pressed pumice blocks as much as 40 cm in diameter that are ptygmatically folded. These features, together with small amounts of pheno-

AGE AVERAGE MODAL COMPOSITION OF TUFFS (volume %) LAVA FLOWS ve volume)

I

28 p.>.:.'.'.'..:..:. :,..>,::.:',;: . _ _ .;:,:'.~.;;:..;.'~<~:'..':.,'.Q:.. _ _ _. . .: .. .. ... .::: .. . w.'.:B. ... . ....I:.'::HC:,.',[;~M . . . .

I I I I I M I 1 P I B RYAN

P l B U ~ +'+' W A H W A H SPRINGS I w I

D < z

X 0 rn ? -4 u, rn --I rn

H M

-{+ c ESCALANTE # +B+M +S DESERT

I Q l I I I 0 5 10 15 2 0 2 5 3 0 3 5 4 0 4 5

Figure 6. Modal proportions of phenocrysts in tuffs of the Indian Peak volcanic field. Phenocryst proportions are on a whole-rock basis free of lithic fragments, which occur in significant quantities in rhyolite tuffs. Dacite tuffs are stippled bars; rhyolite tuffs and the trachydacite Isom tuff are open bars. Thickness of bars does not represent duration of eruption of deposit. Data on units labeled # &om Dinkel (1969), Kreider (1970), Best and others (1973), Anderson and others (1975), and Grant (1978) are averages of many analyses; modes of other units are based on at least 1,500 points counted in a single thii section supplemented by mineral separations. This single mode is considered representative of the unit judging from examination of hundreds of outcrops. P, plagioclase; B, biotite; Q, quartz; H, hornblende; C, clinopyroxene; 0, orthopyroxene; X, orthopyroxene plus cliopyroxene; M, Fe-Ti oxides, chiefly magnetite; S, sanidine; T, titanite. Plus signs precede accessory phases. Most rocks contain trace amounts of apatite and zircon. The right side of the figure shows the relative volumes of rhyolite and andesite lava flows in and near the caldera complex.

BEST AND OTHERS

crysts of plagioclase and two pyroxenes, suggest a relatively high temperature of emplacement.

The Isom is the most distinctive unit in the Indian Peak volcanic field. It is the only tuff in which pyroxenes rather than biotite and hom- blende dominate as mafic phases; the Isom pos- sess the same mineral assemblage as most of the andesitic lavas. Apatite commonly occurs as inclusions in pyroxene, and zircon is sparse.

Andesitic lavas that erupted late in the evolution of the volcanic field underlie the Isom tuff in several localities, mostly in the southern part of the caldera complex. These two-pyroxene (and locally hornblende-bearing) andesites have been included in the Isom Formation. In the southern Needle Range, lavas of this age have compositional characteristics transitional to the distinctive Isom tuff, including high Zr and K 2 0 concentrations (Fig. 8).

COMPOSITIONAL RELATIONS

During its 5-m.y. lifetime, the magma system beneath the developing Indian Peak caldera complex sporadically erupted rhyolite and andesite lavas and cyclically erupted much larger volumes of rhyolite and still more voluminous dacite ash flows. Modal and bulk chemical compositions of these rocks provide some general insights into the nature and evolution of the magma system.

Elemental concentrations were determined by X-ray fluorescence spectrometry at BYU using the Norrish and Hutton (1969) method of sample preparation and data reduction. Major elements were determined on fused glass disks and minor and trace elements on undiluted pressed rock powders. Samples were analyzed in dupli- cate and tied to calibration lines of several NIM and USGS rock standards. Total iron is reported as Fe203. Estimates of precision are shown in Table 1.

With few exceptions, the compositional variety of rocks in the Indian Peak volcanic field (subalkaline, metaluminous andesite to rhyolite) is broadly similar to those in other continental magmatic arcs that developed over subducting oceanic lithosphere (Ewart, 1979). In terms of the lack of Fe enrichment, relatively low Ti02 concentrations (less than I%), and position on a normative plagioclase versus A1203 diagram (Irvine and Baragar, 1971), they are calc- alkaline; the rock series is calcic to calc-alkalic as defined by Peacock (1931). Except for some andesitic lavas, both holocrystalline and glassy rocks of the Indian Peak volcanic field are de- cidedly potassic and fall in the high-K field of Ewart (1979). On the tectonic discrimination diagrams of Pearce and others (1984), which are based on Rb, Nb, and Y abundances, the rhyolites and dacites also show their similarity to subduction-related volcanic-arc granites world-

Figure 7. IUGS classitication (Le Bas and others, 1986) of volcanic rocks from the Indian Peak volcanic field. AN analyses have been recalculated to 100% volatile free. Triangles represent andesites of all ages. Cognate inclusions in tuffs are shown as filed symbols, other samples with letters.

wide. Although perhaps only a result of en- hanced differentiation, Isom tuffs show within- plate tendencies because of high Rb and Nb abundances. Rocks in the Indian Peak volcanic field are similar in composition and mineralogy to those related with other middle Cenozoic caldera complexes in the Great Basin (Best and others, 1989a).

A Andesites E Escalante Desert

C Cottonwood s Spring Wash

W Wah Wah R Ripgut Springs

L Lund I lsom

Andesitic Lavas

Minor volumes of intermediatecomposition lava were extruded from numerous vents in and near the Indian Peak caldera complex during the lifetime of the magma system but chiefly during deposition of the Escalante Desert Formation at -32 Ma. Most are petrographically the same in that they contain phenocrysts of calcic plagioclase, augite, hypersthene, and magnetite in a matrix of the same plus variable amounts of brown glass. A few flows of all ages (such as numbers 9 and 12 in Table 1) also contain hornblende and biotite. Chemically, the intermediatecomposition lava flows are variable and range from basaltic andesite to dacite (Fig. 7). Two flows that plot in the dacite field are not petrographically different from those that lie in the andesite field. Although most are high-K, they range to low-K and no single value of K57.5 (Gill, 1981) exists. This variety suggests that they cannot be comagmatic. The most dktinc- tive variation occurs in MgO concentrations. High and low MgO andesites were extruded throughout the lifetime of the system. Flows with greater than 4.5 wt% MgO (numbers 3,4, 5, and 9 in Table 1) have lower A1203 and P2O5 and higher K20, K20/Na20, and Fe203/Ti02 than other flows at comparable Si02 concentrations. The most primitive flow (no. 5) of the high-MgO type (a basaltic

andesite) extruded well beyond the caldera complex northeast of Crystal Peak, contains phenocrysts of olivine instead of hypersthene. Like volcanic rocks found in other subduction settings, all of the andesitic rocks have negative Nb anomalies (on chondrite normalized diagrams), with the deepest anomalies found in the high MgO lavas.

Rhyolitic Tuffs and Lava Flows

Rhyolitic tuffs and lavas of the caldera complex are a petrographically similar suite of rocks that contain less than about 15% phenocrysts (Fig. 6), chiefly plagioclase (generally andesine), and minor biotite and Fe-Ti oxides; some samples also have hornblende, quartz, clinopyroxene, or sanidine that together generally make up less than 1%. Apatite and zircon are trace constituents.

The rhyolitic rocks are all slightly peralumi- nous and have high K20/Na20 ratios whether or not they are vitrophyres. Although most samples are low-Si02 rhyolites, Si02 contents range from 68% to 77% on an anhydrous basis (Table 2). Only two units can be considered as high- Si02 rhyolites, a lava from the Escalante Desert Formation and the basal Ripgut tuff. These high8i02 rocks, which contain only sparse quartz phenocrysts, differ from other rhyolitic rocks in their lower concentrations of Zr (less than 150 ppm), Fe203, Ti02, and MgO; it is notable, however, that compared to the low- Si02 rhyolites, the high-Si02 rhyolites are not more enriched in the incompatible elements Rb or Nb nor depleted in sanidinecompatible elements such as Ba.

All of the rhyolitic tuffs are normally zoned, both petrographically and chemically. In addition to its phenocryst zonation, the Lamerdorf


A Andesites E Escalante Desert

C Cottonwood S Ryan Spring Wash

I W Wah Wah R Ripgut Springs

L Lund I Isom

Figure 8. Si02 variation diagrams for volcanic rocks from the Indii Peak volcanic field. See caption for Figure 7 for additional information.

-

6 - 5 3 - - n

ON 4- Z -

2 -

Tuff Member of the Escalante Desert Formation shows slight chemical zonation from base to top, with the top being enriched in Ti, Al, Fe, Ca, Na, Sr, and Ba relative to the base. Ryan Spring tuffs show the same sense and magnitude of vertical zonation in composition, as well as in phenocryst proportions and size; in addition, quartz and sanidine disappear upward. The Ripgut tuff also shows phenocryst zonation-in size, abun-

A 8 - A A ~

33 A A

'?+hW 16 8

' ~ 5 R 8

E R

O l ~ ~ ~ I . ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ,

500

400 -

n 300-

k. n - 200-

100-

dance, and type (Fig. 6), with the appearance of hornblende in the upper part as well as the strongest chemical zonation of the rhyolitic tuffs (Figs. 7 and 8).

The younger rhyolitic tuft3 (Ryan Spring and Ripgut) which appear to have been emplaced relatively soon after eruption of the voluminous dacite ash flows, are chemically distinct from the rhyolite lavas and tuffs of the Escalante Desert

rocks differ significantly from early and middle

All three dacite tuff units contain about 40% phenocrysts, more than half of which are zoned calcic andesine (Fig. 6). Other phenocrysts include biotite, hornblende, quartz and minor to trace amounts of clino- and ortho-pyroxene, Fe- Ti oxides (chiefly cubic), zircon, and apatite. The Lund is the only unit that contains titanite and sanidine. Sue of phenocrysts varies within and between units. Overall, the Cottonwood Wash Tuff has the largest phenocrysts, and this character prevails almost to the top of the deposit. In contrast, the Wah Wah Springs outflow member is finest grained (about 2 mm or less) in its lower part and coarsest (5 mm or so) at the top. Crystal size in the Lund is rather uniform throughout.

Cognate inclusions are ubiquitous in the dacite tuff units. Most of these were pumiceous lapilli compacted with the host tuff, but rarely pumice blocks to as much as 30 cm in diameter have been found in weakly welded tuff. The suite of inclusion samples also includes three (nos. 11, 12, and 14) that are weakly vesicular to nonvesicular vitrophyres. Although the cognate relation of all of the inclusions to their host tuff is indicated by similarities of characteristic phenocryst assemblage and bulk composition of each unit, the origin and significance of the nonvesicular inclusions is problematic, and further investigations are underway. Such inclusions have been noted in other ash-flow deposits (for example, Noble and others, 1974).

Although dacite tuft3 are petrographically strikingly different from andesitic lava flows and from rhyolitic tuft3 and lava flows, bulk chemical compositions of dacites slightly overlap the extreme variants of these units and bridge between them to form an overall continuous com-

I II

I E

E~ A A

A S

s A A ~ : A*.* A

o i ~ ~ ~ ~ r ~ ~ ~ ~ ~ ~ ~ n ~ ~ ~ ~ ~ ~ ~ r ~

Formation. The younger, dacite-associated rhyolites have lower concentrations of Zr, Ti02, and K 2 0 and significantly higher CaO at similar Si02 concentrations (Fig. 8). Moreover, silicic rocks of the Escalante Desert Formation extend to lower silica contents; some lavas included in this unit are alkali-rich dacite or trachydacite (Fig. 7). Thus, all of the silicic rocks of the Esca- lank Desert Formation have compositional similarities to the Isom.

As a group, Oligocene calc-alkaline rhyolitic

Miocene rhyolites erupted in the same area (Best and others, 1987~). The genetically distinct, younger highSiOz magmas have abundant phenocrysts of quartz and sanidine, were erupted in small volumes from numerous local vents to form flows, domes, and tuffs, are strongly enriched in F and other incompatible elements, and have higher Fe/Mg ratios (Christiansen and others, 1986).

Dacite Tuffs

8 - ~

6 - - 5 3 - 4- 8 - I

2 -

A

A - A

' AA% AA

v. 1pP I E E g V E

0 1 0 s ~ . - - ~ r ~ ~ b - 8 ~ 8 . . ~ . . ~ ~

1.2-

n 0 . 8 - i Y 3 - 0.6- N 0

0.4-

0.2 - 0

- A

A A ~ A

A , A I I I A L* El

???.WE E

' W s k S

E R

, , , , , , , , , , , , , I , . , , . , ,

BEST AND OTHERS

Thus, compared with dacite-associated rhyolitic rocks of the Ryan Spring and Ripgut For- mations, Isom magmas were erupted at higher temperature (see also next section), contain pyroxenes instead of biotite as major mafic phenocrysts, and have higher alkali contents. Nonetheless, compared to rhyolite l ava of the Escalante Desert Formation whose extrusion initiated the eruptive lifetime of the Indian Peak magma system, the Isom has similar high Zr and low Fe203/Ti02 and CaO/A1203 ratios. Per- haps the genetic conditions which led to these distinctive compositions were nearly reproduced at the beginning and close of eruptive activity of the system. Following the suggestion of Meen (1987), it is possible that high-pressure fractionation of pyroxene from a more mafic parent led to substantial enrichment of K and incompatible elements without a commensurate increase in Si02. Clinopyroxene-dominant fractionation could have p r o d u d the low CaO/A1203 ratios in the differentiates.

positional spectrum that ranges from the low MgO andesites to the low Zr rhyolites (Table 2 and Fig. 8). Part of the chemical variation in the

contain higher elemental concentrations of Ti, Al, Fe, Mg, Ca, P, and Sr, and lower of Si, Na, and K. Microprobe analyses (W. P. Nash, 1985,

dacite tuffs may be a consequence of mechanical fractionation of crystals and lighter glass during eruption and emplacement of the crystal-rich ash-flows (Sparks and Huang, 1980). The volume of far-traveled ash can be a significant, even equivalent, volume compared to the associated ash-flow deposits. G. A. Izett (1987, oral commun.) has found beds of ash near the base of the

written commun.) of glass in cognate inclusions from the base of the outflow and from the top of the intracaldera Wah Wah Springs tuff show that both glasses are rhyolitic with low Fe/Mg ratios and no analytically significant differences. Thus, inclusion compositions suggest some small, but stratigraphically systematic, zonation in the Wah Wah Springs ash flows. Further evaluation of chemical variation in these large- volume dacite deposits must include additional comparisons of intra- and extracaldera tuffi and cognate inclusions, especially their phase com-

Arikaree Group in Nebraska that have age and crystal composition appropriate for the Needles Range dacite tuffs. Some sort of glass/phenocryst fractionation probably explains the differences between analyses 3/4 gnd 7/8 which positions (compare with Grunder and Boden,

1987). Compositions of cognate inclusions of the

Cottonwood Wash and Lund tuffs are distinct

represent matrix/pumice pairs from the Cot- tonwood Wash and Wah Wah Springs tuffs. Elemental enrichments (Ti, Al, Fe, Mg, Ca, Sr, P, and Zr) and depletions (Si, K, and Rb) are for most elements from each other and from

those in the Wah Wah Springs (Fig. 8; Table 3). Moreover, based on bulk compositions of tuff and inclusions, the Lund is the least evolved of

consistent with the accumulation of plagioclase, zircon, apatite, and mafic minerals. Linear variation trends in the dacites (Fig. 8) suggest andesitic and crustal silicic materials were both the dacites. Nonetheless, it has the most quartz Generalities Regarding P-T-X in the

Magma System important in the development of the magmas. Scott and others (1971) report initial 8 7 ~ r / 8 6 ~ r ratios that range from 0.7088 to 0.7099 for pla- gioclases from dacitic tuffs of the Needles Range

of the three, contains sanidine, a smaller propor- tion of mafic phenocrysts, and no free pyroxene. As total abundance of phenocrysts is about the same in all the dacites, slight but significant dif-

Comparisons of the phenocryst assemblages found in rocks of the Indian Peak field with experimentally determined phase relations for similar compositions allow some inferences to be drawn about the intensive parameters of

Group which are consistent with a substantial ferences in intensive parameters, rather than differences in extent of crystallization or composition, must have prevailed in the dacite magma bodies.

crustal component in these rocks. In light of the contention of Whitney and

Stormer (1985) that large-volume dacite tuf i erupted magmas. The experiments of Johnson and Rutherford

(1989) conducted on the Fish Canyon Tuff are are compositionally uniform, it is important to

Isom Trachydacite Tuff examine the evidence for zonation in the dacites of the Indian Peak volcanic field. Contrasts be- probably most applicable to the Indian Peak da- tween immobile-element concentrations in basal Compared to other tuffs in the volcanic field

at the same Si02 content, the trachytic Isom Formation has higher concentrations of Ti02, Zr, Ba, Rb, and K20, and much lower CaO concentrations. In spite of the high alkali content, the Isom rocks are not peralkaline and by strict IUGS usage (Le Bas and others, 1986),

cites, particularly the Lund, which has the same phenocryst assemblage of quartz, plagioclase, sanidine, biotite, hornblende, ilmenite, magne-

vitrophyres and immediately overlying devitrified tuffs of a single deposit (nos. 1 and 2, and 5 and 6) could reflect a slight vertical zonation or tite, and titanite. Experiments show that this

mineral assemblage is stable at relatively low PH,o/P,d, less than 0.5, and at relatively low temperatures of about 760 "C. The assemblage,

simply mechanical fractionation of crystals and glass during eruption and emplacement. Alter- natively, differences between more mobile elements could have developed during hydration most samples are trachydacite because of their

high normative quartz content. Moderate Fe/Mg ratios place them in the calc-alkaline field (Fig. 7). Thus, the Isom tufh are substan- tially different from trachytes erupted from younger caldera complexes in the Great hasin which have higher Fe/Mg ratios and are commonly peralkaline (Novak and Mahood, 1986);

however, is not particularly pressure sensitive, that is, stable from 2 to 5 kb, but measured aluminum concentrations in hornblende in the

and devitrification of the glass. A better test for compositional uniformity is a

comparison of cognate inclusions from within a single deposit. Bulk chemical compositions of

Lund suggest a pressure of crystallization of about 2.5 kb using the calibration of Johnson and Rutherford (1989). The presence of the oxygen-buffering assemblage titanite-magnetite- quartz-clinopyroxene-hornblende in the Lund

pumice lumps from the base and top of the outflow tuff member of the Wah Wah Springs Formation (nos. 8 and 9 in Table 3 and Fig. 8) are sufficiently different in some elements (Ti, however, the Isom is similar to trachytic rocks

from the middle Tertiary Trans-Pews volcanic field (Henry and others, 1988).

The high Zr concentrations of the Isom de-

tuff is consistent with its crystallization at relatively high fugacity of oxygen, nearly 2 log units above the QFM buffer (Noyes and others,

Fe, and P) to exceed analytical and geological uncertainties. Greater zonation in the Wah Wah Springs tuff is evident in a comparison of inclu- 1983). Such high oxygen fugacity is consistent sions from the outflow and intracaldera deposits. mand a high eruption temperature (Watson and

Harrison, 1983). Magnetite-ilmenite and two- pyroxene geothermometry indicates an equili- bration temperature of about 950 "C.

with the over-all calc-alkaline character of the magmas erupted from the Indian Peak magma Inclusions from the intracaldera member are

significantly and systematically more mafic than those from even the top of the outflow tuff and

system. The experiments of Johnson and Ruth- erford (1989) show phase assemblages which


lack sanidine (as in the Wah Wah Springs and Cottonwood Wash tuffs) are formed at temperatures of 800 to 900 OC. Nusbaum (1988) reports Fe-Ti oxide temperatures for the Wah Wah Springs of 800 to 880 OC and oxygen fugacity of about 2 log unit above QFM.

We noted earlier that the intracaldera tuff of the Wah Wah Springs Formation commonly has slightly more quartz than does the outflow tuff. According to the experiments of Clemens and Wall (1981) and Naney (1983) which show positive dP/dT slopes for the quartz phase boundary, a slightly greater pressure and/or lower water content could account for this difference. Either of these differences is compatible with a somewhat deeper source for the intracaldera Wah Wah Springs tuff. The less evolved chemical composition but greater quartz content of the Lund tuff could be explained by the same relative shift in P and water content, or a lower eruption temperature as indicated above.

Rhyolitic rocks of the Indian Peak caldera complex have near-liquidus phenocryst assemblages dominated by plagioclase and biotite with little or no hornblende, sanidine, or quartz, even though concentrations of Si02 and K20 are high, as in the lower Ripgut tuff. The experiments of Naney (1983) and Clemens and Wall (1981) suggest that this assemblage requires water concentrations in the magma in excess of 3 to 4 wt%. The expansion of the quartz stability field relative to plagioclase with increasing pressure documented by Tuttle and Bowen (1958), Clemens and Wall (1981), and Naney (1983), also limits the crystallization of the quartz-poor rhyolites of the Indian Peak magma system to relatively low pressures, less than 4 kb.

The two-pyroxene plagioclase assemblage of the Isom tuff was reproduced by Naney's (1983) experiments with a low-silica granite at temperatures which range from 900 to 980 OC (at 8 kb) and at 820 to 920 OC (at 2 kb) at water concentrations less than 3.5% in the magma. The experiments of Johnson and Rutherford (1989) suggest that the Isom assemblage is stable at low pressures (<2 kb) and temperatures of about 900 OC. These suggestions of high temperature are in accord with the geological evidence and with the relatively high temperatures from mineral geothermometers.

DISCUSSION AND CONCLUSIONS

Important conclusions that can be drawn from the foregoing information include the following.

1. The Indian Peak caldera complex is a cluster of four nested calderas and two poorly de-

fined sources covering a present area of about 80 x 120 km. This caldera complex and related volcanic rocks are the surface manifestation of an underlying, major magmatic system that vented on the order of 10,000 km3 of magma about 32 to 27 Ma.

2. Eruptions from this open magma system, which included discrete magma chambers whose size, shape, and constituents changed with time, yielded a variety of andesites, dacites, rhyolites, and trachydacite rocks that cannot have been strictly comagmatic, that is, derived from a single magma body. These rocks are calc- alkaline with low to moderate Fe/Mg ratios, low TiOz, inferred high oxygen fugacities, and, in the andesitic rocks, sizable Nb depletions. With the exception of some andesites, all are potassic with K20>Na20. Although also showing many calc-alkaline characteristics, low-silica rhyolites and a trachydacite ash-flow with distinctive enrichments of incompatible elements were erupted, respectively, early and late in the evolution of the magma system.

3. In contrast to the Marysvale and San Juan volcanic fields that are marginal to the Colorado Plateaus, no protracted extrusion of andesitic lava flows preceded ash-flow eruptions from the Indian Peak magma system in the Great Basin. Only trivial extrusions (an estimated few percent of the volume of silicic tuff) occurred during its 5-m.y. lifetime. Chemically variable two-pyroxene andesitic magmas erupted throughout the history of the system from vents both within and outside the caldera cluster. Although small in volume, these extrusions are considered to mani- fest a mantle input produced above a subducting slab of lithosphere. Their variable and evolved compositions, however, must reflect differentiation in an open system, including variable mafic magma recharge, crustal contamina- tion, crystal fractionation, and mixing with silicic crustal melts. The greatest volumes of andesite lava flows were extruded early and late in the life of the system. A reduced mid-life rate of andesite eruption could reflect the difficulty of ascent through overlying less dense silicic magma bodies as the magma system matured.

4. The voluminous dacite tuffs and the trachydacite Isom tuff were derived from sources that migrated southward through time (Fig. 2). This transgression mimics in rate and direction the pattern of volcanic activity in the entire Great Basin (Cross and Pilger, 1978; Best and others, 1989a) and suggests that sites of magma generation or eruption shifted in re- sponse to regional dictates.

5. The history of the Indian Peak magma system is marked by cyclic eruptions of petrograph-

ically similar, compositionally zoned, lithic-rich rhyolites and more voluminous, crystal-rich dacites and a final trachydacite ash flow. Rather similar conditions of magma generation and evolution, as well as extent of crystallization prior to eruption, were recurrently achieved in the magma system.

6. In detail, however, conditions of rhyolite and dacite magma generation differed somewhat through the lifetime of the system. For example, Escalante Desert silicic rocks are chemically distinct from younger rhyolites, and modes of the Ryan Spring tuffs show their affinity to the Wah Wah Springs dacite (phenocryst assemblages which lack sanidine), whereas the Ripgut shows similarities to the Lund (both have sanidine and titanite). The erupted Lund magma apparently partially crystallized under different conditions from the other dacites.

7. Eruption of Ryan Spring and Ripgut rhyolite magma appears to have followed closely in time after the voluminous dacite eruptions, rather than the reverse, because epiclastic deposits overlie each of the rhyolite deposits but not the dacites. The oldest rhyolitic magmas from the Indian Peak system, forming the chemically distinctive Escalante Desert Formation, were associated with no precursory dacite, possibly because only enough mantlederived thermal power had been inserted into the crustal system to generate a quasi-bimodal rhyolite-andesite association.

8. It is improbable that rhyolitic magma existed as a capping differentiate of a compositionally stratified magma body immediately before eruption of dacite ash flows. This conclusion follows from (a) the absence of zoned rhyolite to dacite cooling units; (b) inappropriate time rela- tionships (Escalante Desert rhyolitic magmas were erupted about 2 m.y. before the succeeding large-volume dacite of the Cottonwood Wash Tuff, and the occurrence of epiclastic deposits suggest a hiatus between eruption of rhyolite and following dacite); and (c) the difficulty of withdrawing thousands of cubic kilometers of dacite magma from a chamber without tapping any rhyolite magma which hypothetically capped a zoned chamber. Vigorous convection in dacite magma chambers could have pre- cluded the existence of gravitationally stable caps of rhyolite.

9. Distinctive enrichments in TiOz, alkalies, and especially Zr occur in rhyolitic and trachytic rocks formed during waxing and waning stages of the Indian Peak magma system compared to rhyolitic rocks of its mid-life. These differences reflect differences in magma source conditions or subsequent diversification processes.


10. The denouement of the Indian Peak magma system is associated with the migration of magmatism to the south and waning mantle thermal input into the system. Magmas erupted from the "temporal flank" of the waning magma system include the trachydacite Isom tuff and andesitic lavas. The distinctive characteristics of the Isom (high concentrations of alkalies and incompatible elements for its Si02 content, high temperature, and anhydrous mineral assemblage) suggest that it was derived by fractionation of andesitic magmas.

Following a hiatus of at least 3 m.y. after eruption of the Isom magmas from the Indian Peak system, the Caliente caldera complex (Rowley and Siders, 1988; Best and others, 1989a), centered about 40 km to the south (Fig. l), came to life, marking the continued southward migration of volcanism.

ACKNOWLEDGMENTS

The senior author is especially indebted to two friends, Lehi F. Hintze and Thomas A. Steven. Lehi fmt introduced me to the wonder- ful world of Needles Range tuffs two decades ago through an established program of roving BYU summer field geology courses that has produced much of the data reported here; he has been a constant unwavering fountain of help and understanding. Tom has been a most patient and provocative mentor in tuff things, as well as an invaluable liaison in the catacombs of the Denver Federal Center. Many other geologists helped in many ways; they include Ernie Ander- son, John Anderson, Gary Axen, John Bartley, Clark Blake, C. G. "Skip" Cunningham, Bill Dickinson, Allen Glazner, Kerry Grant, Sherm Grommt, Jeff Keith, Bart Kowallis, Marge MacLachlan, Ted McKee, Hal Morris, Bill Nash, Fred Nelson, Keith Rigby, Peter Rowley, M. Shafiqullah, Dan Shawe, Ralph Shuey, Wanda Taylor, Dave Tingey, Chuck Thorman, Jim Whitney, and Julie Barrott Willis. Financial support was provided by the U.S. Geological Survey, Brigham Young University, in part through a Karl G. Maeser Research Award to the senior author, and National Science Founda- tion Grants EAR-8604195 and EAR-8618323. Thorough reviews by Allen Glazner, Richard Hardyman, Thomas Moyer, Peter Rowley, Tom Steven, and anonymous reviewers improved early versions of this manuscript.

REFERENCES CITED

Anderson. J. J., Rowley, P. D., Red, R. J., and Navin, A.E.M., 1975, Cenc- loic geology of southwestern hi@ plateaus of Utah: Geological Society of America S p d Paper 160, p. 88.

Best, M. G., 1987, Geologic map of the area between Hamlin Valley and Esdante Dcsa lron County, Utah: U.S. Geological Survey Miscel- laneous lnvutigation Series Map 1-1774, scale 1:50,000.

But, M. G., and Grant, S. K., 1987, Stratigraphy of the volcanic Oligocene N d e s Range Group in southwestern UUh: U.S. Geological Survey Rofessional Paper 1433A, p. 3-28.

Bes~ M. G., Shuey. R. T.. Culrey, C. F., and Grant, S. K., 1973, Stratigraphic relauonsofmcmbersofthc Needles Rangc Formauon at type local~t~a m southwervrn Uuh Gmlo@cal Soocty of Amcnca Bullcun, v. 84. p. 3269-3278.

But, M. G., McKee, E. H., and Damon, P. E., 1980, Space-timecomposition patfern of late Cenozoic mafic volcanism, southwestern Utah and ad- joining a m : American Journal of Science, v. 280, p. 1035-1050.

But, M. G., Grant, S. K., Cleary, J., Hutsinpiller, A,, and Saunden. D. M., 19874 Geologic map ofthe Indian Peak Range, Beaver and lron Coun- tics, Utah. U.S. Geological Survey Miscellamons lnveaigation Series Map 1-1795, scale 1:50,000.

BesL M. G.. Hintle. L. F.. and Holms R. D.. 198%. Gmlonic mao of the soulhern ~ i u n t a r " Homc ~ . n & and northem ~ndia i ~eak '~angc . Beaver Couav. Uuh. U.S. Gmloaical Survey Mlcceilaneous I n v w a - tion Series blip 1.17%. d e 1 : ~ 0 0 0 .

Best, M. G., Mehnen, H. H., Keith, J. D., and N-r, C. W., 1987% Miocene magmatism and tectonism in and near the southern Wah Wah Moun- tains, southwestern UUh: U.S. Gmlogical Survey Professional Paper 14338.

Best, M. G., Chriniansen, E. H., Deino, A. L., Gromml, C. S., McKee, E. H.. and Noble, D. C., 19894 Late Eocene through Miocene volcanism in the Great Basin of the western United States: New Mexiw Bureau of Mines and Gmlogy Spcd.l Publication (in press).

Best, M. G., Tah, M., Kowallis, B. J., W i i J. B.. and &st, V., L989b, Geologic map of northern White Rodr MountaiasHamlin Valley area, &aver County, Utah and Linwln County, Nevada: U.S. Geological Survey MisccUanmus lnvutigations Series Map 1-1881.

Christiamcn, E. H., Sheridan, M. F., and Bun, D. M., 1986, The geology and gemhemistry of Cenozoic topaz rhyolites from the western United States: Gedogical Society of America Spefial Paper 205,82 p.

Clemens, J. D., and W d , V. J., 1981, Origin and c v d k a t i o n of some perduminous (Stype) g~ni t ic msgmas: Canadian Mineralogis v. 19, p. Ill-131.

Coney, P. J., and Harms, T. A,, 1984, Cordilleran metamorphic mre wm- plexes: Cenozoic extensional relics of M m i c mmpreaion: Geology, v. 12, p. 5 s 5 5 4 .

Crm, T. A., and Pilger, R. H., Jr., 1978, Coosvaiots on absolute motion and plate interaction inferred from Cenozoic igneous activity in the westem United States: American Journal of Science, v. 278, p. 865-902.

Dinkel, T. R., 1%9, Vertical and lated variations in a welded 1uffW.S. thesis]: Rolla, Mimuri, Univndty of Mimuri 49 p.

Ekren, E. B., Mclntyre, D. H.. and Beonen, E. H., 1984. High-temperature. large-volume lavalike, ash-flow tuffs without calderas in southwestern Idaho: U.S. Geological Survey Professional Paper I272 76 p.

Ewart, A,. 1979, A review of the mineralogy and chemistry of Teninry-Recent dacitic, latitic, rhyolitic, and related salic volcanic rocks, in Baker, F.. ed.. Troodhjemites, daciteg and related m c k Amnerdam, the Nether- lands, Elsevier, p. 13-121.

Gill J. B., 1981, Orogenic andesites and plate tectonics: M n , West Germany, Springn-Verlag, 390 p.

Grant, S. K., 1978, Stratigraphic relations of the Esralante Dcsn Formation near Lund, Utah: Brigham Young University Geology Studies, v. 25, D. 27-30.

run&, A L., and Boden, D. R., 1987, Comment on "Magmatic cond~uonr of the Fkb Canyon Tuff,central San Juan volcan~c ficld, Colorado" by Whimev and Stormcr (1985) Journal of Pclmloav. v. 28. D. 737-746.

Henry, C. D., brio+ J. G., i d myth, R. C., 1988, Chemical 6nd thermal zonation in a mildly alkaline magma sys(em, Infemito caldera, Trans Peas, Texas: C o n m i o m to Mineralogy and Peuology, v. 98. p. 19&211.

Hildreth, W., 1981. Gradients in silicic magma chambers: Implications for lithospheric magmatism: Journal of Geophysical Research, v. 86, p. 10,153-10,192.

Irvine, T. N., and Baragar, W.R.A., 1971, A guide to the chemical ckification of the wrnmon volcanic rocb. Canadian Journal of Earth Science, v. 8, p. 523-548.

Johoson, M. C., and Rutherford, M. J., 1989, Experimentally determined wnditions in the Fish Canyon Tuff, Colorado, magma chamber: Jour- nal of Pmlogy (in prea).

Kmder, J. E., 1970, The Lund ash-flow tuff W.S. thesis]: Rolls, Missouri, Univmitv of Mimuri. 73 D.

Lr Bas, M. J., ie Maw. R. W .'~trccke~uo. A,, and Zenenm. 8.. 1986. A chemical ddwficauon of volcaruc rocb bared oo thc total dkali-uhca diagnm: Journal of Petrology, v. 27, p. 745-750.

Lipman, P. W., 1975, Evolution of the Platoro caldera mmplex and related volcanic rocks, southeastern San Juan Mountains, Cdorado: U.S. G e logical Survey Professional Paper 852, I28 p.

- 1984, The roots of ash-flow calderas in western North America: Win- dows into the tops of granitic batholiths: Journal of Geophysical Re- sarch, v. 89, p. 8801-8841.

Loucks, M. D., Tingey, D. G.. But. M. G., Christiansen, E. H.. and Hinue, L. F.. 1989. Geologic map of the Fortification Range. Linwln and White Pine Counties. Nevada: U.S. Geological Survey Miscellaneous Invutigations Series Map 1.1666, r r l e I:50,000.

Meen, J. K., 1987. Formation ofshoshonites from calcalkaline basalt magmas: Geochemical and experimental mnstraints fmm the type locality: Con- triiutions to Mineralogy and Petrology, v. 97, p. 333-351.

Nancy, M. T., 1983. Phirc-equilibria of rock-forming femmagnesian silicatu in granitic system American Journal of Science. v. 283. p. 993-1033.

Noble, D. C., Slemmons, D. 8.. Korringa, M. K., Dickinson, W. R.. Al-Rawi, Y., and McKee. E. H., 1974, Eureka Valley Tuff, eatcentral California acd adjacent Nevada: Geology, v. 2. p. 139-142.

N o d . K., and Hunon. J. T., 1%9, An accurate X-ray spenrographic method for the analysis of a wide range of geologic samples: Geochimica et Casmochimica A& v. 6. n. 90-99. ~~ ~ . ..

Novak. S. W., and Mahood, G. A., 1986, Rise and fall of a Wlt-trachyte- rhvolite m r n a svstem at the Kane Springs Wash caldera: Contribu- tidns to ine era lo& and Petrology, v. 94, p.352-373.

Noyes, H. J., Wones, D. R., and Frey, F. A., 1983, A tale of two plutons: Petrographic and mineralogic wnstrainw on the peuogenuis ofthe Red Lake and Eagle Peak plutons, central Sierra Nevada, California: Journal of Geoloev. v. 91. D. 353-379.

Nusbaum. R ~1.1988, hiagmatc ~mpllcauons from msncral and glars g m chcmlstry for thc outflow mcmbcr of the Wah Wah Spnngs Formauon. Utah and Ncvada EOS (Amencan Gm~hvncal Unlon Transactions). . . v. 69, p. 525.

Peacock. M. A.. 1931, Classification of igneous mcL series: Journal of Geology. v. 39, p. 54-67.

Pearce, J. A., H a d , N.B.W.,and Tindle, A. G., 1984. Traceelementdirrimi- nation dimm for the tstonic internretation of granitic rocks Journal of ~etml&, v. 25, p. 9 5 9 8 3 .

'

R d , J. C., Marvin, R. F., and Nacrer, C. W., 1984, Calderas and ash flow tuffs of the Mogollon Mountains, southwestern New Mexiw: Journal of Geophysical Restarch, v. 89, p. 8713-8732.

Rowley. P. D., and Siden, M. A,, 1988, Miocene calderas of the Caliente caldera wmplex, Nevada-Utah: EOS (American Geophysical Union Tramndions), v. 69, no. 44, p. 1508.

Rowley, P. D., Steven, T. A.. Andemn, J. J., and Cunningham. C. G., 1979. Cenozoic stratigraphic and structural framework of southwestern Utzh: U.S. Gedogical Survey Professional Paper 1149.22 p.

Sargent, K. A., and RoggensacL, K., 1984, Map showing oulcmps of pre- Quaternary ash-flow tuffs and volcaniclastic rocks, Basin and Range province, Nevada: Water-Rmurca Investigations Repon 83-41 19-E.

Scon, R. B.. Nesbitt, R. W., Dasch, E. J., and Armstrong. R. L., 1971, A strontium isotope evolution model for Cenmoic magma genesis, eastern Great Basin, U.S.A.: Bulletin Volcanologique, v. 35, p. 1-26.

Sparks, R. S., and Huang, T. C., 1980, The volcanological signilicance of deep-sea ash layers associated with ignimhrites: Geological Magadne, v. 117, p. 425436.

Steven, T. A,, and Lipman, P. W., 1976, Calderas of the San Juan volcanic field, southwestern Colorado: U.S. Geological Survey Professional Paper 958,35 p.

Steven, T. A., Rowley, P. D., and Cunningharn, C. G.. 1984, Calderas of the Marysvale volcanic field, west central Utah: Journal of Geophysical Research, v. 89, p. 8751-8764.

Stewart. J. H., 1980, Geology of Nevada: Nevada Bureau of Mines and Gml- ogy Spedal Publication 4,136 p.

Stewan, J. H.. and Carlson. J. E.. 1976. Cenozoic mcks of Nevada: Nevada Bweau of M ~ n a and Geology Map 52

Taylor, W J . tn press. Spatla1 and temporal relauons of Cenozo~c %olcamm and extension in the North Pahmc and Seaman Ranas. eastern Ne- vada. m Wcrnrkc. B P. ed. Tentary cxtensnonal tectonirm near thc latatude of Las Vqas. Nevada timlwcal Sooety of Amenca Memoir

Tuttle, 0. F., and ow en,-^. L.. 1958, rigi in of granite & light of experimental studies in the system NaAISi,08-KAISi30,Si0,-HP Geological So- ciety of America Memoir 74, 153 p.

Watson. E. B., and Hanison. T. M.. 1983, Zirwn saturation revisited: Tempera- ture and mmposition effects in a variety of crustal magma types: Earth and Planetary Sdence Lrtteq v. 64, p. 295-M4.

Well$ R. E., and Heller, P. L., 1988. The relative wnbibution of accretion. shear, and extension to Cenowic tectonic rotation in the Pacific Northwest: Geological Society of America Bulletin, v. 100. p. 325-338.

Wbitney, J. A,, and Stormer, J. A., 1985, Mineralogy, petrology, and magmatic muditions from the Fish Canyon Tuff, central San Juan Volcanic Field, Colorado: Journal of Petrology, v. 26, p. 726-762.

Willis, J. B., Best, M. G., Kowallis, B. J.. and Besg V. C., 1987, Preliminary geologic map of northern W i h n C m k range. Linwln County, Ne- vada: U.S. Geological Survey Miscellaneous Field Studis Map MF- 1971. scale l:50.000.

Wun. S L. 1986. R e g o d wmlauon of extension Lrstions m Cordilleran mecamomh~c wrc mmdcxa. Geolw. v. 14. D. 828-830

Zim, I., shuey,k., and ~irb;, J. R., Jr., 1976, ~ e & m ~ e t i c map of Utah: U.S. Geological Survey Geophysical lnvatiptions Map GP-907.

Zieq I., Gilbert, F. P., and Kirby, J. R., 1978, Aeromagnetic map of Nevada: U.S. Geological Survey Geophysical lnvatigations Map GP-922.

Printed in U.S.A.

Oligocene caldera complex and calc-alkaline tuffs and...

Documents

Transcript of Oligocene caldera complex and calc-alkaline tuffs and...