Evidence of a collision between the Yucatán Block and Mexico in the Miocene

Abstract

S U M M A R YWe present the evidence for an anomalous southwest-dipping slab in southern Mexico. The main evidence comes from a clear receiver function image along a seismic line across the Isthmus of Tehuantepec and is also supported by a previous global tomographic model. The slab dips at 35 • , is approximately 250 km in length and appears to truncate the Cocos slab at about 120 km depth. We hypothesize that the slab was created by subduction of oceanic lithosphere prior to the collision of the Yucatán Block with Mexico at approximately 12 Ma. This scenario would explain the Chiapas Fold and Thrust Belt as the product of this collision, and its age constrains the date of the event to be in the Miocene.

Distribution of hydrous minerals in the subduction system beneath Mexico

“…Locations of the converted phases at top and bottom of the subducted oceanic crust are indicated as red and blue dots, respectively. These locations for central Mexico beneath the active arc are confirmed by the predicted arrivals from the synthetic seismograms based on the ray-based algorithm , and the locations for southern Mexico are by the predicted arrivals from the two-dimensional RF finite-difference modeling (Kim et al, 2011). 300 km from the Middle America Trench Kim et al, 2010Kim et al, , 2012. Although the estimate of the thickness slightly trades off with its velocity, the thickness of the oceanic crust is constrained at 7-8 km at the shallow-to-flat regime near the Pacific coast by the finite-difference modeling Kim et al, 2010) with a priori constraint of 5-7 km thick oceanic crust at the trench (Shor and Fisher, 1961).…”

“…Fewer stations and thus a smaller dataset for MASE above the TMVB reflects difficulties in picking phases corresponding to the top and bottom of the oceanic crust in the steeply dipping geometry (see Section 2.2). We also note that the seismic multiples in the teleseismic data recorded from both MASE and VEOX arrays are very weak in our high-frequency passband (Kim et al, 2010(Kim et al, , 2011, and so it is difficult to clearly identify reverberated phases and measure timings of both forward conversions and reverberations to recover Vp/Vs of the subducted low-velocity oceanic crust as done by Audet et al (2009) for Cascadia subduction zone. In particular, the reverberations at the steep interface ( 4 501) cause converted phases to travel over larger lateral offsets, and consequently are beyond the array (MacKenzie et al, 2010;Kim et al, 2012).…”

“…In comparison with the RFs processed in the frequency domain, the iterative deconvolution shows several desirable qualities, and thus for inversion, we only use the quality RF dataset from the iterative deconvolution in time domain for following reasons. The time domain deconvolution provides the highest resolution for sharply constraining the geometry of the slab beneath Mexico (Kim et al, 2010(Kim et al, , 2011, and returns more stable results from noisy stations, especially those above the TMVB. Also, the high frequency content (0.01-1.0 Hz) in the data and iterative time domain deconvolution minimize waveform interference caused by the timing and amplitude variation due to slab dip and slowness.…”

“…The upper-left inset illustrates teleseismic earthquake locations used in the analysis (MASE, 28 events: blue; VEOX, 45 events: green). The upper-right inset shows geometries of the subducted oceanic crust in central (a) and southern (b) Mexico, constrained by the receiver functions Kim et al, 2010Kim et al, , 2011Kim et al, , 2012Melgar and Pé rez-Campos, 2010). dipping oceanic crust in central and southern Mexico from the RF amplitude inversion (Kim et al, 2010), we compare the differences in slab properties between central and southern Mexico, and discuss inferred relationships to the volcanic arc process in the overriding plate and overall plate boundary evolution in Mexico.…”

“…Two black boxes outline the Zitá cuaro-Valle de Bravo (ZVB) and the Sierra Chichiná utzin Volcanic Field (SCVF) in central Mexico (Cervantes and Wallace, 2003;Johnson et al, 2009). Gray lines depict isodepth contours of the subducted Cocos plate beneath the North American plate based on local seismicity and teleseismic receiver functions Kim et al, 2010Kim et al, , 2011Melgar and Pé rez-Campos, 2010). The upper-left inset illustrates teleseismic earthquake locations used in the analysis (MASE, 28 events: blue; VEOX, 45 events: green).…”