Thomas A. Steven

Petrologic evolution of the San Juan volcanic field, southwestern Colorado: Pb and Sr isotope evidence

Abstract Two distinct suites of igneous rocks occur within the San Juan volcanic field: an Oligocene suite of predominantly intermediate-composition lavas and breccias, with associated silicic differentiates erupted mainly as ash-flow tuffs, and Neocene-Pliocene bimodal suite of silicic rhyolites and mafic alkalic lavas. The Oligocene volcanism, probably related to subduction along the western margin of the American plate, has chemical and isotopic characteristics indicative of complex interactions with Precambrian cratonic lithosphere. It also appears to record the rise, differentiation, and crystallization of a large composite batholith beneath the San Juan field. The earliest intermediate-composition lavas and breccias have major- and minor-element compositional patterns indicative of high-pressure fractionation and are relatively nonradiogenic in both Pb and Sr, suggesting significant interaction with lower crust of the American plate. The more silicic ash-flow tuffs show compositional evidence of low-pressure fractional crystallization and are more radiogenic in Pb and Sr — features thought to indicate significant shallow residency for the magmas and interaction with upper crust. Especially radiogenic Pb-isotope compositions of some of these rocks may reflect interactions between the magmas and convecting meteoric water rich in leached Pb, a process thought to have been even more important in forming associated hydrothermal ore deposits. Ore leads tend to be more radiogenic than associated rock leads. Many of the Miocene-Pliocene basaltic lavas seem to be mantle-derived lavas, similar to those of oceanic islands, but some anomalous xenocrystic basaltic andesites, containing relatively nonradiogenic lead, may have been slightly contaminated by lower crustal components. Rhyolitic lavas and intrusions of the bimodal suite are also nonradiogenic in Pb and Sr, in comparison with the Oligocene rhyolites, and do not appear to have interacted with Precambrian upper crust, probably because they erupted largely through the subvolcanic batholith. The Miocene-Pliocene rhyolites are best interpreted as partial melts of lower crust, with the thermal energy to initiate magma generation provided by concurrent basaltic volcanism.

Volcanic History of the San Juan Mountains, Colorado, as Indicated by Potassium–Argon Dating

Volcanic rocks in the San Juan Mountains constitute the largest erosional remnant of a once nearly continuous volcanic field that extended over much of the southern Rocky Mountains and adjacent areas in Oligocene and later time. Recent regional studies have shown that the gross petrologic evolution throughout the San Juan remnant of this field was relatively simple, with initial intermediate lavas and breccias, followed closely in time by more silicic ash-flow tuffs, and ending with a bimodal association of basalt and rhyolite. More limited data from other remnants of the original field indicate a similar evolution. In the San Juan field, voluminous early lavas and breccias—mainly alkali andesite, rhyodacite, and mafic quartz latite—were erupted from numerous scattered central volcanoes onto an eroded tectonically stable terrane. They formed mostly during the interval 35 to 30 m.y. ago, but some probably were erupted earlier and others up to several million years later. About 30 m.y. ago, major volcanic activity changed to explosive ash-flow eruptions of quartz latite and low-silica rhyolite that persisted until about 26 m.y. ago. Source areas for the ash flows are marked by large calderas in the central and western San Juan Mountains. Two groups of lavas and associated rocks of intermediate composition intertongue with the ash-flow sequence: (1) quartz latitic lavas that were erupted in and adjacent to caldera structures and are genetically related to the ash-flow activity; and (2) other, generally more mafic lavas and related rocks that are widely distributed without evident structural relation to the ash-flow eruptive centers. The second group apparently represents a continuation of the early intermediate activity into the period of major ash-flow eruption. In the early Miocene the character of volcanism changed notably. Whereas the Oligocene volcanics are predominantly intermediate lavas and related silicic differentiates, the younger rocks are largely a bimodal association of basalt and high-silica alkali rhyolite. Basalt and minor rhyolite were erupted intermittently through the Miocene and Pliocene, and at one time formed a widespread thin veneer over the older volcanic terrane. The marked contrast between the Oligocene intermediate to low-silica rhyolitic magmas and the later basaltic and rhyolitic magmas implies either different conditions of magma generation or processes of differentiation for the two suites. This petrologic change coincides approximately in time with nearby development of the Rio Grande depression, a major rift that is the local expression of widespread late Cenozoic crustal extension. Whatever the cause of the petrologic change, the progression from predominantly intermediate to bimodal basalt-rhyolite volcanism, approximately concurrent with initiation of late Tertiary crustal extension, appears characteristic of Cenozoic volcanism for much of the western interior United States.

Multiple ages of mid-Tertiary mineralization and alteration in the western San Juan Mountains, Colorado

Ore deposits in the western San Juan Mountains formed intermittently in middle to late Tertiary time, from about 30 to 10 m.y. ago, during essentially the same span as that of associated igneous activity, as indicated by 31 new K-Ar and fission-track ages. Mineralization occurred recurrently during the waning stages of evolution of several precaldera central volcanoes and also after formation of the Uncompahgre, San Juan, Silverton, and Lake City calderas. The richest ore deposits are associated with structures of the Silverton caldera, but they were emplaced 5 to 15 m.y. after the caldera formed. This mineralization appears genetically unrelated to evolution of the caldera and its associated magmatic system but seems closely related in space and time to volumetrically minor intrusions of quartz-bearing silicic porphyry. This association is common in many mining districts in the Rocky Mountains region.

Origin and structural implications of upper Miocene rhyolites in Kingston Canyon, Piute County, Utah

Kingston Canyon is one of the deepest antecedent canyons in the High Plateaus subprovince of the Colorado Plateaus. Here the East Fork of the Sevier River flows westward transversely across the gently east tilted Sevier Plateau, which is developed on a basin-range fault block uplifted more than 1,500 m along the Sevier fault zone on the west. Upper Tertiary rhyolites, uncommon in southwestern Utah, occur both on the northern rim and in the bottom of Kingston Canyon. Those on the northern rim consist of lava flows and volcanic domes of the rhyolite of Forshea Mountain, dated by K-Ar methods at 7.6 m.y. old. Those in the bottom of Kingston Canyon, the rhyolite of Phonolite Hill, are especially well exposed and provide spectacular examples of a pyroclastic cone whose base is about at river level and a steep-sided volcanic dome emplaced into and through these deposits. The pyroclastic deposits, formerly 500 or more metres thick, consist of airfall, mudflow, and ash-flow(?) material of rhyolite and foreign lithic fragments, especially olivine basalt. The dome consists of flow-banded, mostly devitrified rhyolite as much as 500 m thick; it has been dated by K-Ar methods at 5.4 m.y. In addition to the rhyolites, a dome and lava-flow complex, the rhyodacite of Dry Lake, occurs near the northern rim and is considered to postdate the rhyolite of Forshea Mountain and predate the rhyolite of Phonolite Hill. The rhyolite of Forshea Mountain was deposited near basin-range faults, before the uplift of the Sevier Plateau and before the cutting of Kingston Canyon. Before uplift, a river flowed across the site of the present Sevier Plateau toward the east-southeast and perhaps also across the Awapa and Aquarius Plateaus to the east. The rhyodacite of Dry Lake was deposited during uplift and perhaps before canyon cutting. During uplift, the river maintained itself and cut Kingston Canyon. The rhyolite of Phonolite Hill was deposited in this canyon, blocking the river flow, which probably formed new outlets to the east. The Awapa and Aquarius Plateaus later were uplifted along faults, disrupting the eastern part of the river segment. The topography then took on its present appearance, and drainage was re-established through Kingston Canyon. There has been little deepening since the reopening of Kingston Canyon.

Rhyolites in the Gillies Hill–Woodtick Hill area, Beaver County, Utah

The rhyolite of Gillies Hill forms a cluster of rounded hills between Beaver basin and Cove Fort basin in southwest-central Utah. These rocks were erupted as a series of viscous lava flows, volcanic domes, and minor pyroclastic rocks from centers localized along and near the main fault separating the volcanic rocks in the Marysvale volcanic field to the east from batholithic rocks in the Mineral Mountains to the west. Faulting began in middle Miocene time before eruption of the rhyolite of Gillies Hill and continued episodically into Pleistocene time. Potassium-argon dating indicates that the rhyolite of Gillies Hill formed from a rapid sequence of eruptions about 9.1 m.y. ago. The rhyolite of Gillies Hill is made up of an older high-silica suite (SiO 2 >75%) consisting of numerous short stubby flows and of a younger low-silica suite (SiO 2 = 70%) represented by a single large volcanic dome. The high-silica suite is believed to have formed by eruptions from the top of a compositionally zoned magma chamber. The younger low-silica suite came either from a separate magma chamber or from a significantly different level within the same magma chamber. The proximity of sources for both suites suggests eruptions from different levels within a single source chamber, perhaps from different vertically stacked convection cells.