MODERN CHIAPANECAN VOLCANIC ARC

from THE ORIGIN OF MODERN CHIAPANECAN VOLCANIC ARC IN SOUTHERN MEXICO INFERRED FROM THERMAL MODELS
by Vlad C. Manea and Marina Manea, Seismological Laboratory 252-21, CalTech, Pasadena, CA 91125, USA, 2http://www.geo.mtu.edu/~raman/papers2/Manea_Chiapas_2005.pdf

THE ORIGIN OF MODERN CHIAPANECAN VOLCANIC ARC IN SOUTHERN MEXICO INFERRED FROM THERMAL MODELS
Southern Mexico is a very interesting area where the subducting Cocos slab drastically changes its geometry: from a flat slab in Central Mexico 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 2D steady state thermo-mechanical model explains the calc-alkaline volcanism by high temperature (1300� C) in the mantle wedge just beneath the Central America volcanic arc and 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 Miocene Chiapanecan extinct 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 Mexico. The initiation of flat subduction in Central Mexico 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 Chich�n volcano are proposed to be related to the arrival of the highly serpentinized Tehuantepec Ridge beneath 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.

INTRODUCTION

One of the most interesting volcanic areas in Mexico is located in the southern part, in Chiapas. Here, the volcanism is characterized by two Neogene volcanic arcs: the ancestral Miocene Sierra Madre arc and the modern Chiapanecan volcanic arc (MCVA). The Miocene Sierra Madre arc was abandoned between 9 and 3 Ma, when the MCVA was formed (Damon and Montesinos, 1978). There are a series of characteristics that make this new MCVA special. First, the position of the arc is well inside the continent, at distances of 300 - 350 km from the Middle America trench (MAT). The subducting Cocos slab lies 200 km below the MCVA, whereas the majority of subduction related volcanic arcs are located where the slab is at 100 km depth. Second, although there is a little variation of the slab dip beneath Chiapas (40 - 45�) (Rebollar et al., 1999), the MCVA shows an oblique trending with the MAT of ~ 30�. The MCVA holds typical continental arc calc-alkaline magmas (Damon and Montesinos, 1978), but with one exception: El Chichon volcano. This volcano presents a transitional magma signature from calc-alkaline to adakitic-like type (Macias et al., 2003; De Ignacio et al., 2003). This character was associated with melting of the Cocos slab just beneath at 200 km by De Ignacio et al. (2003). The triple junction proximity between the North America, Cocos and Caribbean plates was proposed by Nixon (1982) as a cause for the alkaline volcanism of El Chichon. Other authors suggest that this volcano is associated with the subduction of the Cocos plate (Luhr et al., 1984; Bevis and Isacks, 1984). The northwesternmost closest neighbor of the MCVA is the isolated Tuxtla Volcanic Front (TVF). On the basis of trace element studies, the Na-alkaline products of TVF are explained by Nelson et al. (1995) to be generated from mantle and continental crust melts contaminated by fluids or melts expelled from the subducted lithosphere. To the south, after a gap of 100 km, the closest neighbor of the MCVA is the calc-alkaline volcano Tacana, which belongs to the Central American volcanic arc (CAVA) (Garcia-Palomo et al., 2004). Although, the slab dip, age and convergence rate of the Cocos slab beneath the CAVA in Guatemala are very similar to those beneath Chiapas, the two active volcanic arcs are in completely different locations. Therefore, it is likely that an external perturbation source of the mantle wedge flow is responsible for this anomalous MCVA. Anisotropy in the form of share wave splitting has been observed in many subduction zones (Kuril, Japan, Tonga, Izu-Bonin), and appear to be related with the asthenospheric flows produced by present-day plate motion (Fischer et al., 2000). Since, thermal and flow models in the mantle wedge can give us advance insights on the geodynamic processes beneath the volcanic arcs, we compute four 2D thermal models normal to MAT, three in southern Mexico and one in Guatemala, using the
numerical approach proposed by Manea et al. (2005-C).

The relationship between the long-term evolution of the subducting Cocos slab and the origin of MCVA has never been addressed so far. In order to explain the unusual MCVA, we propose a geodynamic model mainly related with the initiation of flat subduction in early-middle Miocene (Ferrari et al., 1999) beneath Central Mexico.

TECTONIC SETTINGS

The geometry of Wadati-Benioff zone shows that the Cocos slab increases its dip from 25� just NW of TR in Oaxaca, to 40� southeast of TR in Chiapas (Rebollar et al., 1999, Bravo et al., 2004). Then the slab slightly increases its dip to 45� beneath the northern edge of the CAVA. Also the maximum depth extent of the intraslab earthquakes increases from 120 km in southern Oaxaca (Bravo et al., 2004) to 240 km in southern Chiapas (Rebollar et al., 1999).
Offshore, a very large linear feature, the Tehuantepec ridge (TR), represents a tectonic border between two different regions on the Cocos plate (Manea et al., 2004-A). Across TR, there is an age jump of the Cocos slab at the MAT of ~ 10 Ma, defining two different thermal gradients across the oceanic lithosphere (Figure 2). The seafloor age just northwest of TR is ~ 16 Ma (Kanjorsky, 2003; Klitgord and Mammerickx, 1982), then ~ 26 Ma (Manea et al., 2004-B) in front of the Chiapas state, and then ~ 23 Ma in Guatemala (Manea et al, 2004-B). The rate of convergence between the Cocos plate and the North America and Caribbean plates varies from 6.5 cm/yr in Oaxaca to 7 cm/yr in Chiapas from NUVEL 1A model (DeMets et al., 1994).
Onshore, a very important tectonic feature is represented by the triple junction between three plates: North America, Cocos and Caribbean plates. The contact between NA and Caribbean plates is not very clear. Guzman and Meneses (2000) proposed that this contact is actually a diffuse boundary represented by a fault jog system.
There are two volcanic arcs in Chiapas: the ancient extinct Sierra Madre Miocene arc which runs parallel to the coast, and the MCVA further inland. The active volcanic arc is tilted by ~ 30� in respect with MAT. Damon and Montesinos (1978) show that the Miocene Sierra Madre arc was abandoned between 9 and 3 Ma, then the MCVA was born as a consequence of a reorganization of the Cocos plate. Using isotopic ages (Ar-Ar method) for Central Mexico, Ferrari et al. (1999) show that the position of volcanism in Central Mexico changed during the last 25 Ma, revealing that the Cocos slab became subhorizontal ~ 20 Ma ago.

THERMAL MODELS AND MAGMATIC PRODUCTS

In this study we developed four thermal models constrained by seismic data (Rebollar et al., 1999; Engdahl and Villasenor, 2002), convergence rate (DeMets et al., 1994) and oceanic plate age at the MAT (Manea et al., 2004-B; Kanjorsky, 2003; Klitgord and Mammerickx, 1982). The first profile was chosen just above TR because of the subhorizontal shape of the subducting Cocos slab. Because of the temperature-dependent viscosity, this model shows high temperature (~ 1300� C) beneath the volcanic arc. Using the phase diagram for basalt and harzburgite (Hacker et al., 2003), we can see that the slab geotherm intersects the dehydration melting solidus at ~ 70 km, implying that melting of the basaltic crust might take place at this depth. This is in good agreement with the conclusion of Nelson et al. (1995), who show that a constant ratio of Ba/Nb is an indicator of magma contaminated by fluids or melts derived from the subducted slab. Also, a vertical P-T profile beneath TVF intersects the wet solidus for peridotite (Wyllie, 1979), this explaining the calc-alkaline signature of the volcanic material (Figure 4B). The second profile is situated just below TR and intersects the MCVA in the proximity of the El Chichon volcano. The modeling results show high temperatures beneath the ancient Miocene volcanic arc rather than beneath the MCVA. The distant MCVA lies ~ 200 km above the slab, where the entire oceanic slab is completely dehydrated. Also, the slab geotherm does not intersect the melting solidus for basalt, therefore with this model we cannot explain the K-alkaline products of the El Chichon volcano. At this stage a 2D steady-state thermal model is inconsistent with the observations, but in the following chapters we will address these issues. The last two models are located at the northern edge of the CAVA . The results show a region with high temperature (~ 1300� C) just beneath the active arc, well beyond the wet peridotite. Also strong slab dehydration (~ 5 wt.%) lowers the melting point of mantle peridotite, and favors calc-alkaline volcanism. The slab geotherm does not intersect the basalt solidus, this being consistent with the pure calc-alkaline character of the CAVA (no Na-alkaline or K-alkaline products).

SPACE-TIME EVOLUTION OF THE COCOS SLAB AND CHIAPANECAN VOLCANISM
One of the key issues in understanding the position through time of a subducting slab is the variation in time of volcanic arcs. Using the distribution of dated rocks of Ferrari et al. (1999) for Central Mexico, and of Damon and Montesinos (1978) for Chiapas, we propose the evolution of the subducting slab presented in Figure 5. Between 25 and 17 Ma ago, the volcanic arc formed an approximatively continuous belt in Central Mexico, Chiapas and Guatemala. Then, between 17 and 12 Ma ago, the CMVB moved inland, suggesting that the subducting slab become subhorizontal. To the south, the rest of the volcanic arc remained parallel with MAT, therefore suggesting that the slab approximately preserved its previous dip. Later, between 12 and 7 Ma, the flattening process of the Cocos slab continued further SE, while the CAVA retreated SE. Between 7 and 3 Ma ago, the CAVA continued to retreat SE and the volcanic activity close to the Chiapas coast ceased completely. The TVF was born during this time period. The last episode of this scenario took place between 3 and 0 Ma, and is represented by the onset of the MCVA.
Since the slab dip, age and convergence rate of the Cocos slab beneath the Central American volcanic front in Guatemala are very similar to those beneath Chiapas, why are the two active volcanic arcs so different, and why are both not active? Therefore, to answer this question we looked for an external perturbation source responsible for the anomalous Chiapanecan volcanism. We propose that an external perturbation came from the time evolution of the subducting slab beneath Central Mexico. The onset of the flat slab just above TR, in Central Mexico, would have produced a strong mantle wedge inflow in the neighboring region to the SE. This intake would have migrated over time, creating the tilted MCVA. The isotopic ages of the MCVA decrease from NW to SE, suggesting that the arc was propagating through time from NW to SE. If such a model can explain the position of the distant tilted MCVA, there is still one question that has to be addressed: why has the Sierra Madre volcanic arc ceased? To answer this question we look to the implication of the 3D temperature distribution of this region. The flat cold incoming oceanic lithosphere (profile A-A'' )is in contact with the hot mantle wedge just ~120 km SE. The flat slab temperature is ~600� C, while the mantle wedge temperature is ~1300� C. Therefore there is a continuous cooling of the mantle wedge beneath the ancient Chiapanecan volcanic arc. If a mantle wedge tip is cooled down, then a very interesting phenomena might appear. From the phase diagram for harzburgite we can see that during the
cooling, the mantle wedge enters into the serpentine stability field (~ <600� C). The location of serpentinized mantle wedge tip is critical because it may control the down-dip coupling between the slab and the overlying material (Manea et al., 2004). Very low shear-wave velocities in the cold forearc mantle have been discovered in the southern Cascadia subduction zone (Bostock et al., 2002). This is evidence of a highly hydrated and serpentinized material in the forearc region. The same conditions should also be expected in the Chiapas subduction zone.

THE ORIGIN OF MODERN CHIAPANECAN VOLCANIC ARC IN SOUTHERN MEXICO INFERRED FROM THERMAL MODELS
Next
South Mexico