PETROGENESIS AND GEODYNAMIC SIGNIFICANCE OF SILICIC VOLCANISM IN THE WESTERN TRANS-MEXICAN VOLCANIC BELT

Silicic volcanism in the western Trans-Mexican Volcanic Belt (WTMVB) was defined a Pliocene ignimbrite flare-up associated with the rifting of the Jalisco block from mainland Mexico (Frey et al., 2007; GSAB). With the integration of new and published geochronologic, geochemical, and isotope data we revise this interpretation and propose a new petrogenetic model. The oldest silicic volcanism consists of large silicic domes and minor pyroclastic flows (~370 km³) emplaced to the north of Guadalajara above a thick succession of ~11 to 8.7 Ma basaltic lavas, which yielded Ar-Ar and obsidian FT ages of ~7.5 to 5 Ma. Shortly after (4.9 to 2.9 Ma) large amount of rhyolitic lavas and ash flow tuffs (~500 km³) were emplaced in a WNW-ESE trending belt from Guadalajara to Compostela. Rhyolitic domes and flows (~430 km³) were emplaced also in the Pleistocene mostly between Tequila and Guadalajara with the late Pleistocene La Primavera caldera (~35 km³) as the sole explosive volcanic episodes. As a whole, silicic volcanism occurred from Late Miocene to the Pleistocene, and was dominated by dome and lava flows.

Most rhyolites have high LILE/HFSE values and negative spikes at Nb, P and Ti. They also show the same Ba/Nb and K/Rb values and slightly higher Rb/Sr ratios as the 11-8 Ma basalts. Rhyolite Sr isotope data (⁸⁷Sr/⁸⁶Sr init = 0.70371 – 070598) are only slightly more radiogenic than the 11-8 basalts (⁸⁷Sr/⁸⁶Sr init = 0.70349-0.70410), whereas Nd isotope ratios are indistinguishable from them. Sr and Nd isotope ratios of the rhyolites are also similar to the crust nearby, indicating that they can be compatible either with fractional crystallization (FC) of basalts or with crust assimilation/melting. However REE contents are too low to be the result of basalt FC alone. Isotope and REE data can be successfully modeled with an initial crustal melt which subsequently undergone fractional crystallization of feldspar and quartz. Late Miocene slab detachment and subsequent slab rollback produced pulses of mafic magma that were partly trapped in the crust yielding crustal melting. Extensional faulting since the Pliocene favor the eruption of silicic magma as effusive dome and lava flows. Rifting at the boundaries of the Jalisco block is seen as a rollback induced reactivation of crustal structures but is unlike to evolve into a Jalisco microplate.