SYNCHRONOUS 29-19 MA ARC HIATUS, EXHUMATION, SUBDUCTION OF FOREARC, AND BREAKUP OF FARALLON PLATE IN SOUTHWESTERN MEXICO: POSSIBLE CAUSES

The geological record of southern Mexico documents several synchronous events in the upper Oligocene-lower Miocene (29-19 Ma): (1) an apparent hiatus in arc magmatism between the WNW-trending, upper Eocene–lower Oligocene (uE-lO), arc magmatism in the central part of the Sierra Madre del Sur, and the E-W Miocene-Present, Trans-Mexican Volcanic Belt (TMVB): arc magmatism: east of the Veracruz-Oaxaca Line started at ~22 Ma; (2) removal of a wide uE-lO forearc; (3) exhumation of 13-20 km of uE-lO arc along the present day 40-80 km wide coastal belt; (4) breakup of the Farallon Plate into the Guadalupe and other smaller plates. These effects have traditionally been related to eastward displacement of the Chortis block from a position off southwestern Mexico, however, at 30 Ma the Chortis block would have lain opposite the Gulf of Tehuantepec, not off southwestern Mexico. We propose an alternative explanation: subduction of the uE-lO forearc that may be linked with the tomographically-defined, 200-350 km high, vertical step in the subducted plates, which connects two gently dipping segments, one deep and the other shallow extending to the Acapulco Trench. Initiation of vertical subduction would have extinguished arc magmatism in the Sierra Madre del Sur, which was reinitiated along the TMVB at ~19 Ma following reestablishment of gently dipping subduction at ~24 Ma: this allows only ~5 my for removal of a wedge-shaped uE-lO forearc. If the forearc was relatively cool continental lithosphere, it may have been underplated beneath the overriding plate, however, a relatively warm zone is revealed in tomography north of the step that may be related to magma rising beneath the TMVB. On the other hand, if the subducted forearc was floored by oceanic lithosphere it may have been converted to eclogite and been transported with the subducting slab to great depths. The cause of subduction of the forearc does not appear to be related to either a change in the rate of convergence, which doubled after 20 Ma, or the age of the subducting oceanic lithosphere. It may be due to either collision of anomalous oceanic lithosphere, possibly a large oceanic plateau, or breakup of the shallow dipping Farallon Plate into a vertically dipping Guadalupe Plate, which decoupled the forearc from the arc.