Alteration and remelting of nascent oceanic crust during continental rupture: Evidence from zircon geochemistry of rhyolites and xenoliths from the Salton Trough, California

Abstract

Rhyolite lavas and xenoliths from the Salton Sea geothermal field (Southern California) provide insights into crustal compositions and processes during continental rupture and incipient formation of oceanic crust. Salton Buttes rhyolite lavas contain xenoliths that include granophyres, fine-grained altered rhyolites (“felsite”), and amphibole-bearing basalts. Zircon is present in lavas and xenoliths, surprisingly even in the basaltic xenoliths, where it occurs in plagioclase-rich regions interpreted as pockets of crystallized partial melt. Zircons in the xenoliths are exclusively Late Pleistocene–Holocene in age and lack evidence for inheritance. U–Th isochron ages are: 20.5_− 1.2^+ 1.2 ka (granophyres), 18.3_− 3.5^+ 3.6 ka (felsite), 30.1_− 12.4^+ 14.1 ka and 9.2_− 6.6^+ 7.0 ka (basalts; all errors 1σ). The dominant zircon population in the rhyolite lavas yielded U–Th ages between ∼ 18 and 10 ka, with few pre-Quaternary xenocrysts present. δ¹⁸O_zircon values are lower than typical crustal basement values, thus ruling out rhyolite genesis by melting of continental crust. Moreover, δ¹⁸O_zircon values are ∼ 0.5–1.0‰ lower than compositions achievable by zircon crystallization from residual melt in equilibrium with unaltered mid-ocean ridge basalt, suggesting that basaltic crust and silicic plutons in the subsurface of the Salton Sea geothermal field isotopically exchanged with meteoric waters. This is evidence for deep-reaching hydrothermal circulation and indicates rhyolite genesis by episodic remelting of altered basalts instead of fractional crystallization of unaltered basaltic magma.

Introduction

Understanding the thermo-mechanical dynamics of lithospheric deformation during rifting is essential for assessing the resource and hazard potentials associated with continental rupture. The Gulf of California–Salton Trough is an outstanding example of ongoing rapid horizontal and vertical deformation, fast sediment accumulation, high heat flow, magmatism and sea-floor spreading that accompanies fragmentation of the North American continental margin [1], [2]. Onshore, the region hosts extensive geothermal systems (e.g., Salton Sea, Cerro Prieto), and there is potential for off-shore hydrocarbon and mineral deposits [3]. The main kinematic stages associated with continental rifting and embryonic formation of oceanic crust in the Gulf of California–Salton Trough are well understood [4], [5], [6]: (1) cessation of Pacific plate subduction at ∼ 12 Ma; (2) eastward stepping of the Pacific–North American plate boundary by ∼ 250 km; (3) reorganization of the new plate boundary from a transtensional deformation zone into en-echelon pull-apart basins separated by transform fault systems; and (4) maturation of these pull-apart basins into spreading centers.

The Salton Trough region is the subaerial extension of the Gulf of California rift system, but despite intense geophysical coverage by gravity, seismic and heat-flow surveys, ambiguity exists as to the extent of thinned basement versus young mafic crust and metamorphosed rift-basin sediment [7], [8], [9], [10]. Synrift magmatic rocks from the surface and subsurface of the Salton Trough are therefore essential in identifying crustal compositions. Previous studies of igneous rocks in the Salton Sea area recognize a bimodal assemblage of rhyolite lavas and basalt xenoliths, with the basalts resembling mid-ocean ridge basalts (MORB) from the Alarcon and East Pacific Rise spreading centers [11], [12]. In addition, the rhyolites contain granitic xenoliths. The origins of these rhyolites and the granitic xenoliths are unclear. Different origins have been proposed, including extreme fractionation of basalt, partial melting of hydrous peridotite, partial melting of granitic basement, and assimilation of metasedimentary crust by differentiated basalt [11], [12], each having different implications for the thermal and compositional state of the crust.

In contrast to previous geochemical studies that mostly utilized whole-rock analysis, we explore the U–Th–O isotope and trace element compositions of individual zircon crystals from Salton Buttes lavas and xenoliths in order to delimit the timing and origins of magmatism. Zircon is particularly well-suited due to its stability and the slow diffusion of Th, U, and O relative to most other minerals (e.g., [13]) so that it preserves a record of its origins even in the high temperature and fluid flow environment of an active magmatic–geothermal system. We performed ∼ 50 high spatial resolution ion microprobe analyses of zircon from surface lavas and xenoliths that indicate Late Pleistocene to Early Holocene magmatism between ∼ 30 and 10 ka. We also find magmatic zircon δ¹⁸O compositions that rule out an origin via anatexis of typical continental crustal sources or differentiation of unaltered mantle melts. Instead, our results reveal two previously unrecognized processes for the origin and composition of Salton Trough silicic magmas: (1) deep-reaching circulation of meteoric waters causing compositional alteration of juvenile mafic crust, and (2) episodic remelting of newly accreted oceanic crust to generate rhyolites.