Terry C. Wallace

Subduction and collision processes in the Central Andes constrained by converted seismic phases

The Central Andes are the Earths highest mountain belt formed by ocean–continent collision. Most of this uplift is thought to have occurred in the past 20 Myr, owing mainly to thickening of the continental crust, dominated by tectonic shortening. Here we use P-to-S (compressional-to-shear) converted teleseismic waves observed on several temporary networks in the Central Andes to image the deep structure associated with these tectonic processes. We find that the Moho (the Mohorovičić discontinuity—generally thought to separate crust from mantle) ranges from a depth of 75 km under the Altiplano plateau to 50 km beneath the 4-km-high Puna plateau. This relatively thin crust below such a high-elevation region indicates that thinning of the lithospheric mantle may have contributed to the uplift of the Puna plateau. We have also imaged the subducted crust of the Nazca oceanic plate down to 120 km depth, where it becomes invisible to converted teleseismic waves, probably owing to completion of the gabbro–eclogite transformation; this is direct evidence for the presence of kinetically delayed metamorphic reactions in subducting plates. Most of the intermediate-depth seismicity in the subducting plate stops at 120 km depth as well, suggesting a relation with this transformation. We see an intracrustal low-velocity zone, 10–20 km thick, below the entire Altiplano and Puna plateaux, which we interpret as a zone of continuing metamorphism and partial melting that decouples upper-crustal imbrication from lower-crustal thickening.

Crustal-thickness variations in the central Andes

We estimated the crustal thickness along an east-west transect across the Andes at lat 20°S and along a north-south transect along the eastern edge of the Altiplano from data recorded on two arrays of portable broadband seismic stations (BANJO and SEDA). Waveforms of deep regional events in the downgoing Nazca slab and teleseismic earthquakes were processed to isolate the P-to-S converted phases from the Moho in order to compute the crustal thickness. We found crustal-thickness variations of nearly 40 km across the Andes. Maximum crustal thicknesses of 70–74 km under the Western Cordillera and the Eastern Cordillera thin to 32–38 km 200 km east of the Andes in the Chaco Plain. The central Altiplano at 20°S has crustal thicknesses of 60 to 65 km. The crust also appears to thicken from north (16°S, 55–60 km) to south (20°S, 70–74 km) along the Eastern Cordillera. The Subandean zone crust has intermediate thicknesses of 43 to 47 km. Crustal-thickness predictions for the Andes based on Airy-type isostatic behavior show remarkable overall correlation with observed crustal thickness in the regions of high elevation. In contrast, at the boundary between the Eastern Cordillera and the Subandean zone and in the Chaco Plain, the crust is thinner than predicted, suggesting that the crust in these regions is supported in part by the flexural rigidity of a strong lithosphere. With additional constraints, we conclude that the observation of Airy-type isostasy is consistent with thickening associated with compressional shortening of a weak lithosphere squeezed between the stronger lithosphere of the subducting Nazca plate and the cratonic lithosphere of the Brazilian craton.

Crust and mantle structure across the Basin and Range‐Colorado Plateau boundary at 37°N latitude and implications for Cenozoic extensional mechanism

We present new evidence on the seismic velocity and density of the crust and upper mantle along a 200-km-long transect across the eastern Basin and Range and western Colorado Plateau at 37°N latitude. Receiver functions computed from the P waveforms recorded with 10 portable broadband stations deployed along the transect were used to estimate crustal thickness variations. The crust is 30-35 km thick within the eastern Basin and Range and increases over a distance of ∼100 km at the western edge of the Colorado Plateau, reaching a maximum of approximately 45 km east of the Hurricane fault. The timing of crustal multiples within the reciever functions were used to estimate the V p /V s of the crust along the profile, and we found that the western Colorado Plateau crust is characterized by a high Poissons ratio (0.28-0.29) indicative of a crust with an average mafic composition. We estimated the upper mantle lid thickness along our profile based on teleseismic P wave travel times and constraints provided by gravity data. Our data and available geophysical constraints are most consistent with a lithosphere that thickens from an average thickness of 60 km beneath the Basin and Range to 100 km beneath the western Colorado Plateau, although the Basin and Range lithosphere may have significant thickness variations. The thick, strong mafic crust and thicker mantle lid under the Colorado Plateau can account for the relative geologic stability and subdued magmatism of the plateau during Laramide compression and Cenozoic extension compared to surrounding regions. The crustal and lithospheric thinning across the tectonic boundary occurs over a short distance (∼100 km), suggesting it is a geologically young feature produced by a predominantly mechanical response to late Cenozoic extension. Our new lithosphere model at 37°N latitude is consistent with the existence, in early Cenozoic time, of a flat subducted slab at 100 km depth and a relict Sevier-Laramide 50-60 km thick crustal welt, and 60-100% pure shear extension (β values of 1.6-2.0) during the late Cenozoic.

The active tectonics of the eastern Himalayan syntaxis and surrounding regions

Source parameters of 53 moderate-sized earthquakes, obtained from the joint inversion of regional and teleseismic distance long-period body waves, provide the data set for an analysis of the style of deformation and kinematics in the region of the Eastern Himalayan Syntaxis. Focal mechanisms of Eastern Himalayan events show oblique thrust, consistent with the N-NE directed movement of the Indian plate as it underthrusts a boundary that strikes at an oblique angle to the direction of convergence. Earthquakes near the Sagaing fault show strike-slip mechanisms with right-lateral slip. Earthquakes on its northern splays, however, indicate predominant thrusting, evidence that the dextral motion on the Sagaing fault, which accommodates a portion of the lateral motion between India and southeast Asia, terminates in a zone of thrust faulting at the Eastern Himalayan Syntaxis. Remaining motion between India and southeast Asia is accommodated in a zone of distributed shear in east Burma and Yunnan, manifested by strike-slip and oblique normal faulting, east-west extension, crustal thinning, and clockwise rotation of crustal blocks. We determined strain rates throughout the region with a moment tensor summation using 25 years (modern) and 85 years (modern and historic) of earthquake data. We matched the observed strains with a fifth-order polynomial function, and from this we determined both the velocity field and rotations with respect to a specified region. Velocities calculated relative to south China stationary show that the entire area, extending from 20°N–36°N, within deforming Asia (Yunnan, western Sichuan, and east Tibet), constitutes a distributed dextral shear zone with clockwise rotations up to 1.7°/m.y., maximum in the region of the Eastern Syntaxis proper. Integrated strains across this zone, relative to south China stationary, show 38 mm/yr ± 12mm/yr of north-directed motion at the Himalaya. Remaining plate motion, relative to south China fixed, must be taken up by the underthrusting of India beneath the lesser Himalaya, strike-slip motion on the Sagaing fault, and intraplate NE directed shortening within NE India as well as NE directed shortening within the Eastern Syntaxis proper. 10 mm/yr ± 2 mm/yr of relative right-lateral motion between India and southeast Asia is absorbed in the region between the Sagaing and Red River faults (94°E–100°E). It is the clockwise vorticity (relative to south China) associated with the deformation in Yunnan, east Tibet, and western Sichuan that provides the relative north-directed motion of 38 ± 12 mm/yr at the Himalaya. Not all of the deformation is accommodated in right-lateral shear between India and south China and between east Tibet and south China; velocity gradients exist that are parallel to the trend of the shear zone. Relative to a point within western Sichuan (32°N, 100°E), the velocity field shows that the Yunnan crust is moving S-SE at rates of 8–10 mm/yr. Relative to south China, there is no eastward expulsion of crustal material beyond the eastern margin of the Tibetan plateau.

Rupture Characteristics of the Deep Bolivian Earthquake of 9 June 1994 and the Mechanism of Deep-Focus Earthquakes

The Mw = 8.3 deep (636 kilometers) Bolivian earthquake of 9 June 1994 was the largest deep-focus earthquake ever recorded. Seismic data from permanent stations plus portable instruments in South America show that rupture occurred on a horizontal plane and extended at least 30 by 50 kilometers. Rupture proceeded at 1 to 3 kilometers per second along the down-dip azimuth of the slab and penetrated through more than a third of the slab thickness. This extent is more than three times that expected for a metastable wedge of olivine at the core of the slab, and thus appears to be incompatible with an origin by transformational faulting. These large events may instead represent slip on preserved zones of weakness established in oceanic lithosphere at the Earths surface.

Accretionary tectonics of Burma and the three-dimensional geometry of the Burma subduction zone

The geometry of the Burma Wadati-Benioff zone (WBZ) has been determined by fitting a trend surface parameterized with eight effective degrees of freedom to 184 well-located hypocenters. The dip of this surface, which passes through the middle of the WBZ, varies from about 50° in the north near the eastern Himalayan syntaxis to about 30° in the Bay of Bengal area. The eastern edge of the Indo-Burman ranges closely follows the map projection of the 60 km depth contour of the WBZ. The curvature of the Indo-Burman ranges is controlled by the geometry of the interface between the more steeply dipping part of the Indian plate and the leading edge of the overriding Burma platelet. Shallow earthquakes beneath the Indo-Burman ranges are primarily confined to the underthrusting Indian plate. Their focal mechanisms indicate strike-slip faulting and north-south shortening parallel to the eastern margin of the Indian plate.

Lithospheric‐scale structure across the Bolivian Andes from tomographic images of velocity and attenuation for P and S waves

We have developed a three-dimensional, lithospheric-scale model across the Bolivian Andes at ∼20°S, based on tomographic images of velocity and attenuation for both P and S waves. Observations of travel time and attenuation for this study are from regional, mantle earthquakes in the subducted Nazca plate recorded on a portable, broadband seismic array (Broadband Andean Joint Experiment and Seismic Exploration of the Deep Andes) in Bolivia and Chile. The shallow mantle under the Altiplano from ∼18°S to ∼21°S is high-velocity and moderately high Q (Vp ≈ 8.3,Vs ≈ 4.7, Qp ≈ 500, and Qs ≈ 200), suggesting lithospheric mantle. High-velocity material in the Altiplano extends to a depth of ∼125–150 km. The shallow mantle of the Western Cordillera is characterized by high Vp/Vs (∼1.83), suggesting a correlation between Vp/Vs and arc volcanism. Seismic velocity in the Western Cordillera mantle is, on average, only slightly reduced from global averages; however, velocity and attenuation anomalies are locally strong ( Vp ≈ 7.8, Vs ≈ 4.3, Qp ≈ 200, and Qs ≈ 100), consistent with partial melt conditions. Under the Los Frailes volcanic field, in the Eastern Cordillera, shallow mantle velocity and Q decrease drastically from the neighboring Altiplano ( Vp ≈ 7.8, Vs ≈ 4.3, Qp ≈ 300, Qs ≈ 100); however, high Vp/Vs is not as pervasive as it is in the Western Cordillera. We believe that slab-derived water, and perhaps other volatiles, strongly influence the Western Cordillera, while the Eastern Cordillera low-velocity region is more affected by partial melt and/or compositional changes. Average velocity and Q in the shallow mantle across the Bolivian Andes, where the tomographic images are best resolved, are significantly higher than in most mantle wedge environments where corresponding images are available. This is likely the result of a compressional “back arc” setting in the Andes. This implies that lithospheric shortening and thickening associated with the formation of the Andes has profoundly influenced the shallow mantle structure across the range. Shallow mantle structure is locally influenced by the subduction processes, particularly under the Western Cordillera; however, the differing volcanism and seismic character under the two Cordilleras suggest that the volcanic process in the Eastern Cordillera may be distinct from arc volcanism. Tertiary volcanism in the Eastern Cordillera is located in the region where mantle shortening is suspected to be greatest. Both the timing and location of volcanism are consistent with upward migration of mantle wedge asthenosphere following the removal of over thickened lithosphere.

Active tectonics of the Pamirs and Karakorum

The source parameters of moderate- to large-sized earthquakes and the spatial distribution of earthquake hypocenters are used to investigate the active tectonics and three-dimensional configuration of the subducted lithosphere in the Pamir-Karakorum intracontinental convergence zone. The continental collision has resulted in intense deformation in the upper crust. The deformation is distributed over a broad area and is absorbed by thrust and strike-slip faulting. Along the northern margin of the Pamirs, where the deformation due to continental collision is most intense, the Northern Pamir Thrust and several major fault zones mark the present boundary. Most of the earthquakes that occurred in these fault zones are shallow crustal events with focal mechanism solutions correlated with the tectonic features of the region. Along the Pamir Front, thrust faulting dominates, while on the western and eastern edges of the Pamirs the style of deformation is characterized by oblique thrusting with a component of strike-slip motion. Thrust-type events beneath the Tadjik Depression indicate that both the sedimentary rocks and the basement are involved in shortening. Right-lateral strike-slip motion is observed on the eastern edge of the Pamirs where they border the Tarim Basin and in the Talas-Fergana-Kun Lun fault zone. A southward dipping seismic zone beneath the Pamirs and Karakorum indicates that the Asian lithosphere has been subducted along the Pamir Front to a depth of at least 200 km. This interpretation is consistent with published geological and geophysical data. Focal mechanism solutions of some intermediate-depth events beneath the Pamirs show strike-slip faulting with approximately N-S horizontally oriented P axes, indicating that nearly horizontal compression at the intermediate depth is the predominant mode of deformation. A 90-km-deep event beneath the Karakorum is interpreted as occurring in the subducted Asian lithosphere; the fact that the P axis of this event is oriented parallel to the descending plate suggests downdip compression within the Asian lithosphere. The overall tectonics of the region is interpreted as a consequence of the underthrusting Asian lithosphere being impinged upon by the shallow northward underthrusting Indian lithosphere.

Shear wave anisotropy beneath the Andes from the BANJO, SEDA, and PISCO experiments

We present the results of a detailed shear wave splitting analysis of data collected by three temporary broadband deployments located in central western South America: the Broadband Andean Joint experiment (BANJO), a 1000-km-long east-west line at 20°S, and the Projecto de Investigacion Sismologica de la Cordillera Occidental (PISCO) and Seismic Exploration of the Deep Altiplano (SEDA), deployed several hunderd kilometers north and south of this line. We determined the splitting parameters Φ (fast polarization direction) and δt (splitting delay time) for waves that sample the above- and below-slab regions: teleseismic * KS and S, ScS waves from local deep-focus events, as well as S waves from intermediate-focus events that sample only the above-slab region. All but one of the * KS stacks for the BANJO stations show E-W fast directions with δt varying between 0.4 and 1.5 S. However, for * KS recorded at most of the SEDA and PISCO stations, and for local deep-focus S events north and south of BANJO, there is a rotation of Φ to a more nearly trench parallel direction. The splitting parameters for above-slab paths, determined from events around 200 km deep to western stations, yield small delay times (≤0.3 s) and N-S fast polarization directions. Assuming the anisotropy is limited to the top 400 km of the mantle (olivine stability field), these data suggest the following spatial distribution of anisotropy. For the above-slab component, as one goes from east (where * KS reflects the above-slab component) to west, Φ changes from E-W to N-S, and delay times are substantially reduced. This change may mark the transition from the Brazilian craton to actively deforming (E-W shortening) Andean mantle. We see no evidence for the strain field expected for either corner flow or shear in the mantle wedge associated with relative plate motion. The small delay times for above-slab paths in the west require the existence of significant, spatially varying below-slab anisotropy to explain the * KS results. The implied anisotropic pattern below the slab is not easily explained by a simple model of slab-entrained shear flow beneath the plate. Instead, flow induced by the retrograde motion of the slab, in combination with local structural variations, may provide a better explanation.

The determination of source parameters for small earthquakes from a single, very broadband seismic station

The installation of very broadband seismic stations makes it possible to recover the source parameters of small earthquakes (2.5 < ML < 5.0) which occur at local and regional distances. If the gross crustal structure along the travel path is known, it is possible to use the P, SV and SH displacement waveforms from a single station to determine the seismic moment tensor. Although the details of the crustal structure strongly affect the body waveforms at regional distances, the signature of the seismic source orientation on the waveform is robust at frequencies less than 1–3 Hz. We explore the trade-offs between crustal model, hypocentral depth and filtering for a linear moment tensor inversion procedure. The procedure is tested on two small earthquakes which occurred in the Rio Grande Rift and were recorded at the IRIS/USGS station ANMO. The agreement between the single station moment tensor inversion fault plane parameters and those determined from local first motions is excellent.