Isolating active orogenic wedge deformation in the southern Subandes of Bolivia

Abstract

A new GPS‐derived surface velocity field for the central Andean backarc permits an assessment of orogenic wedge deformation across the southern Subandes of Bolivia, where recent studies suggest that great earthquakes (>Mw 8) are possible. We find that the backarc is not isolated from the main plate boundary seismic cycle. Rather, signals from subduction zone earthquakes contaminate the velocity field at distances greater than 800 km from the Chile trench. Two new wedge‐crossing velocity profiles, corrected for seasonal and earthquake affects, reveal distinct regions that reflect (1) locking of the main plate boundary across the high Andes, (2) the location of and loading rate at the back of orogenic wedge, and (3) an east flank velocity gradient indicative of décollement locking beneath the Subandes. Modeling of the Subandean portions of the profiles indicates along‐strike variations in the décollement locked width (WL) and wedge loading rate; the northern wedge décollement has a WL of ~100 km while accumulating slip at a rate of ~14 mm/yr, whereas the southern wedge has a WL of ~61 km and a slip rate of ~7 mm/yr. When compared to Quaternary estimates of geologic shortening and evidence for Holocene internal wedge deformation, the new GPS‐derived wedge loading rates may indicate that the southern wedge is experiencing a phase of thickening via reactivation of preexisting internal structures. In contrast, we suspect that the northern wedge is undergoing an accretion or widening phase primarily via slip on relatively young thrust‐front faults.

Active transpressional tectonics in the Andean forearc of southern Peru quantified by10Be surface exposure dating of an active fault scarp

“…At the Quaternary time scale, in contrast to the pure compression in Chile, a clear oblique component is distributed in the southern Peruvian forearc along the Purgatorio and Sama‐Calientes faults [ Audin et al , ]. Similarly, the largest rotations derived from GPS data are observed in the forearc of southern Peru [ Bevis and Martel , ; Weiss et al , ]. Based on the comparison of real‐time GPS data with paleomagnetic data, Allmendinger et al .…”

Extension and Dynamics of the Andes Inferred From the 2016 Parina (Huarichancara) Earthquake

“…Both (a) and (b) are in oblique Mercator projection about the Euler pole of the Nazca‐South America convergence taken from DeMets et al (); therefore, slip vectors (SVs) parallel to relative plate motion will be horizontal (McKenzie & Parker, ). (b) The azimuth of shortening in the eastern Andes inferred from selected GPS sites (Kendrick et al, ; Weiss et al, ) and earthquake slip vectors, compared with gradients in the filtered topography (300‐km Gaussian filter) and the elastic block model of Bevis et al (). GPS velocites in (b) are shown relative to stable South America with 1 σ uncertainty ellipses.…”

Rapid Geodetic Shortening Across the Eastern Cordillera of NW Argentina Observed by the Puna‐Andes GPS Array

“…We use the broader set of 136 stations in our realization of the South America fixed reference frame, but we here report on velocities for only the 29 stations between 22°S and 29°S. Some workers have observed an anomalously long‐lasting (>2 years) and large‐magnitude (>15 mm) postseismic signal in the east component of the UNSA time series following the 14 November 2007 M w 7.7 Tocopilla earthquake near Antofagasta, Chile (Weiss et al, ). However, this signal does not affect the time interval under consideration in our study.…”

“…In addition to the thrust‐perpendicular component of the residual velocity field discussed in detail above, there is an appreciable thrust‐parallel component of the velocity field remaining after we subtract the subduction interface‐predicted horizontal velocities from the estimated cGPS velocities in our study area. What has been done previously in similar tectonic settings is a joint inversion of the thrust‐perpendicular and thrust‐parallel components of the velocity field for dip‐slip and strike‐slip parameters, respectively, on a single fold‐thrust belt décollement to account for the entire back‐arc velocity field (Brooks et al, ; Weiss et al, ). This method is appropriate for regions in which the thrust‐parallel component of the velocity field exhibits a roughly arctangent pattern along the profile (Figure b).…”