In southern México, the subducting Cocos slab drastically changes its geometry: from a flat slab in central México to a ∼45° dip angle beneath Chiapas. Also, the currently active volcanic arc, the modern Chiapanecan volcanic arc, is oblique and situated far inland from the Middle America trench, where the slab depth is ∼200 km. In contrast, the Central America volcanic arc is parallel to the Middle America trench, and the slab depth is ∼100 km. A two-dimensional steady-state thermomechanical model explains the calc-alkaline volcanism by high temperature (∼1300 °C) in the mantle wedge just beneath the Central America volcanic arc and the strong dehydration (∼5 wt%) of the Cocos slab. In contrast, the thermal model for the modern Chiapanecan volcanic arc shows high P-T conditions beneath the coast where the extinct Miocene Chiapanecan arc is present, and is therefore unable to offer a reasonable explanation for the origin of the modern Chiapanecan volcanic arc. We propose a model in which the origin of the modern Chiapanecan volcanic arc is related to the space-time evolution of the Cocos slab in central México. The initiation of flat subduction in central México in the middle Miocene would have generated a hot mantle wedge inflow from NW to SE, generating the new modern Chiapanecan volcanic arc. Because of the contact between the hot mantle wedge beneath Chiapas and the proximity of a newly formed cold, flat slab, the previous hot mantle wedge in Chiapas became colder in time, finally leading to the extinction of the Miocene Chiapanecan volcanic arc. The position and the distinct K-alkaline volcanism at El Chichón volcano are proposed to be related to the arrival of the highly serpentinized Tehuantepec Ridge beneath the modern Chiapanecan volcanic arc. The deserpentinization of Tehuantepec Ridge would have released significant amounts of water into the overlying mantle, therefore favoring vigorous melting of the mantle wedge and probably of the slab.

A finite-element method is applied to model the thermal structure of the subducted Pacific plate and overlying mantle wedge beneath the southern part of the Kamchatka peninsula. A numerical scheme solves a system of 2D Navier-Stokes equations and a 2D steady-state heat transfer equation. A model with isoviscous mantle exposed very low temperatures (∼800°C) in the mantle wedge, which cannot account for magma generation below the volcanic belt. Instead, a model with strong temperature-dependent viscosity shows a rise in the temperature in the wedge. At a temperature of more than 1300°C beneath the active volcanic chain, melting of wedge peridotite becomes possible. Although the subducting slab below the Kamchatka peninsula is rather old (ca. 70 Ma), some frictional heating (µ = 0.034) along the interface between the subducting oceanic slab and the overlying Kamchatka peninsula lithosphere would be enough to melt subducted sediments. Dehydration (>5 wt% H 2 O release) occurs in the subducting slab because of metamorphic changes. As a consequence, hydration of the mantle wedge peridotite might produce melt, which may rise to the base of the continental crust as diapirlike blobs. Considering that melting processes in the subducting plate generate most of the volcanic material, we developed a dynamic model that simulates the migration of partially melted buoyant material in the form of blobs in the viscous mantle wedge flow. Blobs with diameters of 0.4–10.0 km rise to the base of the continental lithosphere within 0.002–10 m.y. depending on blob diameter and surrounding viscosity. The thermal structure obtained in the model with temperature-dependent viscosity is used to estimate seismic compressional wave (P-wave) velocity anomalies (referenced to the Preliminary Reference Earth Model) associated with subduction beneath Kamchatka. A low-velocity zone (∼−7% velocity anomaly) is obtained beneath the volcanic belt, and a high-velocity anomaly (∼4%) is obtained for the cold subducted lithosphere. These results agree with seismic tomography results from P-wave arrivals.