Examination of the deep levels of an island arc: Evidence from the Tonsina Ultramafic‐Mafic Assemblage, Tonsina, Alaska

Abstract

The Tonsina ultramafic and mafic assemblage, located in the northern Chugach Mountains along the Trans‐Alaska Crustal Transect (TACT), is interpreted to be the remains of a magma chamber that crystallized at the base of a mature interoceanic island arc. It is one of a number of isolated masses of peridotite and gabbro that comprise the basal part of the Peninsular terrane along the northern fringe of the Chugach Mountains in southern Alaska. These masses, the roots of a mid‐Jurassic island arc, have been uplifted along the north dipping Border Ranges fault system and crop out sporadically for 700 km from Kodiak Island to eastern Alaska. The Tonsina ultramafic‐mafic assemblage is a layered sequence whose basal rocks are dunite and harzburgite (1 km thick) overlain by a narrow zone of websterite and clinopyroxenite. Above these ultramafic rocks is a 5‐ to 6‐km thick layered gabbro that is garnet bearing at its base. Textural relationships and chemical trends throughout the series of dunite, pyroxenite, garnet bearing gabbro and spinel gabbronorite point to an origin by igneous accumulation from a hydrous, tholeiitic parent magma at deep levels, i.e., at the base of a mature island arc before being emplaced along the Border Ranges fault. The rocks were crystallized from a melt at pressures between 9.5 and 11 kbar (28‐ to 33‐km depth) and equilibrated at temperatures of 800° to 90O°C based on phase relationships in the cumulus assemblages, pyroxene thermobarometry, and olivine‐spinel thermometry. Transitional sequences from ultramafic to mafic rocks are characterized by increasingly AL‐rich pyroxenes and spinels (or garnet) and the absence of olivine + plagioclase assemblages. Deep‐level fractionation of Al‐poor phases such as olivine and pyroxene at moderately high pressures and temperatures resulted in concentration of Al in the evolving magma. As crystallization continued, the phases recorded this alumina concentration and the resultant gabbros are Al‐rich. This observed trend supports the theory of high‐alumina basalt generation by fractional crystallization of olivine and pyroxene. Comparison of the Tonsina rocks with Aleutian lower crustal xenoliths as well as other exposures of arc‐related lower crustal magma chambers in Japan, Pakistan, and western North America indicates that this process is a common mechanism for island arc magma evolution at depth.

Thermal structure due to solid-state flow in the mantle wedge beneath arcs

“…(AX [DeBari et al, 1987]). Talkeetna arc section, south-central Alaska: Tonsina area, garnet gabbros a few hundred meters above the Moho (TT [DeBari and Coleman, 1989] and this paper, Figure 3 and Table 1); and Klanelneechina klippe, lower crust (TK, this paper, Figure 3 and Table 1). Petrological estimates of mantle-melt equilibration conditions: NEJF, NE Japan arc front, and NEJC, NE Japan, central arc [Tatsumi et al, 1983]; SOC, southern Oregon Cascades [Elkins Tanton et al, 2001].…”

“…Ellipses with heavy lines are quartz-bearing samples from newly discovered garnet gabbros in the Klanelneechina klippe, about 50 km SW of the Tonsina area. Ellipses with thin lines are for Tonsina area samples, from a thin layer of garnet gabbro immediately above the petrological Moho at the base of the arc section in the Tonsina area [DeBari and Coleman, 1989]; these lack quartz, so the PT estimates are based on a variety of equilibria involving amphibole as well as pyroxenes, plagioclase and garnet. Dashed lines indicate position of temperature/depth sections in Figure 6.…”

“…For the Talkeetna section, DeBari & Coleman estimated metamorphic conditions during garnet formation in gabbros at the base of the crust at ~ 1 GPa, 825-900°C [DeBari and Coleman, 1989]. In Figure 3 and Table 1, we present seven new PT determinations for hornblende-bearing garnet gabbros and garnet gabbronorites from the Talkeetna section made with THERMOCALC v3.1 ( [Holland and Powell, 1988], May 2001 database).…”

High-K ankaramitic melt inclusions and lavas in the Upper Cretaceous Eastern Srednogorie continental arc, Bulgaria: Implications for the genesis of arc shoshonites

“…Recently, such melts have been successfully obtained in melting experiments with variable Ca-rich pyroxene + olivine + amphibole assemblages at 0.5-1 GPa and temperatures of 1175-1350°C (Médard et al, 2004(Médard et al, , 2006. Such assemblages are observed as parts of exposed old arc ultramafic/ mafic cumulate complexes in the Talkeetna area, Alaska (DeBari and Coleman, 1989;DeBari and Sleep, 1991;Kelemen et al, 2003), the Kohistan Terrane of North Pakistan (Khan et al, 1993;Miller and Christensen, 1994;Burg et al, 1998;Ringuette et al, 1999;Jagoutz et al, 2007), and the Ivrea Zone in North Italy (Mehnert, 1975;Rivalenti et al, 1984;Quick et al, 1994). These complexes have been interpreted to be remnants of magma chambers emplaced at the base of the crust, close to the upper mantle-lower crust transition.…”

Early Proterozoic oceanic crust in the northern Colorado Front Range: Implications for crustal growth and initiation of basement faults

“…However, the nature of this crust and the structures along which it was accreted are still poorly understood. Isotopic data suggest a juvenile origin for the Early Proterozoic terranes, with little to no incorporation of Archean crustal material [DePaolo, 1981;Nelson and DePaolo, 1984;Bennett and DePaolo, 1987;Aleinikoff et al, 1993], but recent geochronologic data require the existence of preexisting continental crust [e.g., Selverstone et al, 2000;Hill and Bickford, 2001]. The Proterozoic basement is pervasively fractured parallel to the contact with the Archean craton (Figure 1), suggesting that these structures formed during Early Proterozoic assembly, but few if any of these structures appear to be discrete suture zones.…”