SODIUM-RICH rocks of trondhjemite–tonalite–dacite (TTD) or –granodiorite (TTG) suites form much of Precambrian continental crust1. They are thought to have formed by partial melting of subducted oceanic crust2,3—a process that would have been much more widespread early in Earth history than at present, owing to the higher thermal gradients prevailing at that time4. Phanerozoic TTD suites do exist, however, and seem also to relate to subduction zones5. Defant and Drummond6 proposed that these suites form where young (<25 Myr), hot oceanic lithosphere is subducted and melts, thus locally simulating the conditions that led to widespread crustal growth in the Archaean. Here we describe plutonic and volcanic rocks from the Cordillera Blanca complex in Peru, which have characteristics of the high-Al TTD suite but which were produced above a subduction zone containing a 60-Myr-old slab. We present evidence that the complex formed by partial melting of newly underplated basaltic crust, and argue that this mechanism should be considered more generally as an additional way of generating sodium-rich arc magmas.

Generation of sodium-rich magmas from newly underplated basaltic crust

The enigma of continental plateaus formed in the absence of continental collision is embodied by the Altiplano-Puna, which stretches for 1800 km along the Central Andes and attains a width of 350‐400 km. The plateau correlates spatially and temporally with Andean arc magmatism, but it was uplifted primarily because of crustal thickening produced by horizontal shortening of a thermally softened lithosphere. Nonetheless, known shortening at the surface accounts for only 70‐ 80% of the observed crustal thickening, suggesting that magmatic addition and other processes such as lithospheric thinning, upper mantle hydration, or tectonic underplating may contribute significantly to thickening. Uplift in the region of the Altiplano began around 25 Ma, coincident with increased convergence rate and inferred shallowing of subduction; uplift in the Puna commenced 5‐10 million years later.

The evolution of the altiplano-puna plateau of the central andes

Broadband receiver functions developed from teleseismic P waveforms recorded on the midperiod passband of Regional Seismic Test Network station RSCP are inverted for vertical velocity structure beneath the Cumberland Plateau, Tennessee. The detailed broadband receiver functions are obtained by stacking source-equalizd horizontal components of teleseismic P waveforms. The resulting receiver functions are most sensitive to the shear velocity structure near the station. A time domain inversion routine utilizes the radial receiver function to determine this structure assuming a crustal model parameterized by many thin, flat-lying, homogeneous layers. Lateral changes in structure are identified by examining azimuthal variations in the vertical structure. The results reveal significant rapid lateral changes in the midcrustal structure beneath the station that are interpreted in relation to the origin of the East Continent Gravity High located northeast of RSCP. The results from events arriving from the northeast show a high-velocity midcrustal layer not present in results from the southeast azimuth. This velocity structure can be shown to support the idea that this feature is part of a Keweenawan rift system. Another interesting feature of the derived velocity models is the indication that the crust-mantle boundary beneath the Cumberland Plateau is a thick, probably laminated transition zonemore » between the depths of 40 and 55 km, a result consistent with interpretations of early refraction work in the area.« less

Seismic evidence for an ancient rift beneath the Cumberland Plateau, Tennessee: A detailed analysis of broadband teleseismic P waveforms

Anorthosite massifs, rapakivi granites, and late proterozoic rifting of north America

Rocks of the granulite facies have been proposed as major constituents of the lower continental crust. To evaluate this possibility, compressional and shear wave velocities have been determined to pressures of 10 kb for 10 granulite samples, thus enabling comparisons of seismic data for the lower crust with the velocities and elastic properties of granulite rocks. The samples selected for this study range in composition from granitic to basaltic, with bulk densities of 2.68 to 3.09 g/cm 3 . At 6 kb, compressional ( V p ) and shear ( V s ) wave velocities range from 6.39 to 7.49 km/sec and from 3.36 to 4.25 km/sec, respectively. Velocities in granulite rocks are shown to vary systematically with variations in mineralogical constitution. Both V p and V s , increase with increasing pyroxene, amphibole, and garnet. Velocities increase with an increasing ratio of pyroxene to amphibole in hornblende-granulite subfacies rocks of approximately equivalent chemical compositions. Decreasing quartz content in granulite rocks produces an increase in V p and an accompanying decrease in V s , thereby significantly changing Poisson9s ratio. The range of velocities measured for the granulite samples is similar to the range of seismic velocities reported for the lower continental crust; thus, the hypothesis that granulite rocks are major lower crustal constituents is further strengthened. Furthermore, it is shown that lower crustal composition is extremely variable, and therefore valid discussions of composition must be limited to specific regions where seismic velocities are well known. The use of seismic velocities in estimating lower crustal composition is illustrated for the Canadian Shield in Ontario and Manitoba.

Constitution of the Lower Continental Crust Based on Experimental Studies of Seismic Velocities in Granulite

Summary In 1968, the Carnegie Institution of Washington together with North and South American collaborators carried out a reconnaissance explosion seismic experiment to investigate the apparently highly anomalous crustal structure under the Peru-Bolivia altiplano. The data of this experiment have been reinterpreted by ray-tracing in a spherical Earth so as to fit as closely as possible arrival times, relative amplitudes, cusps, etc., of seismograms displayed in record section. The resultant model confirms the previous average model consisting of three major refractors: the sedimentarymetamorphic layer 49 km thick and 4.54.9 km s-' velocity; the ' granitic ' layer with 6 .O-6.1 km s-' velocity down to 26-30 km depth; and the ' gabbroic ' layer reaching depths of 68-70 km below sea level with 6.8-6.9 km s-' velocity. However, in order to account for relatively large amplitudes in the secondary arrivals with apparent velocities close to the first arrivals, two low-velocity zones are postulated within the crust under the Peru-Bolivia altiplano. In Peru, the shallow and thinner lowvelocityzone with boundaries at 9 km and 12 km depth is between materials of 6.0 km s-' and 6.1 km s-'. The deeper and thicker low-velocity zone with upper and lower bounds at about 30 km and 40 km under Bolivia, and more approximately at 36 km and 46 km under Peru, is embedded in 6.8 and 6.9 km s-' materials. The shallower low-velocity zone is conceivably related to the parent magma of volcanic and intrusive acidic rocks with the deeper low-velocity zone related to the volcanic and intrusive basic rocks in accord with petrological and geochemical findings of Pichler and Zeil in the Andes of northern Chile. The presence of velocity inversions above the 50km depth is also in harmony with the postulated existence of a high electrical conductivity zone shallower than 50 km depth under the Andes as postulated recently by Schmucker to explain magnetic ' day-time fluctuation ' anomalies.

/pdf/crustal-low-velocity-zones-under-the-peru-bolivia-altiplano-4r621xppjq.pdf

Crustal Low-Velocity Zones Under the Peru-Bolivia Altiplano

The source of the geological disturbance responsible for the midcontinent gravity high is here reexamined on the basis of gravimetric, seismological, and geological data in consort, and both the size of the structure and the lithologies responsible are at variance to the results of earlier studies that did not employ long-range, seismological data. Its extension under Lake Superior, implied recently on gravity and seismic grounds and a new seismic inference that flanking areas that extend laterally are typified by gabbroic rather than granitic velocities, infers a disturbance laterally more extensive than that of the high and hence the name used, the central North American rift system. A refined method of analysis of existing long-range crust-upper mantle profile data has made possible the inclusion of this seismic data for the first time. The analysis method rests on the recognition of the existence of a common finite and discrete regional suite of velocities determined from shorter reversed profiles traversing key areas, the assumption of local planarity of interfaces, which then are approximated piecewise, the existence of a functional relationship between compressional velocity and bulk density for crustal materials of the region, and, finally, on the admission that long ‘reversed’ profiles are in the main unreversed and must be treated so that the resulting models simultaneously satisfy the seismic, gravimetric, and geologic data. A high velocity, 6.9 km/sec, forms the ‘core’ of the gravity high imbedded in material of 6.4-km/sec compressional velocity. The inferred density contrast, 0.14 gm/cm3, is substantially smaller than any used previously and results in a 40% increase of volume and an increase in the thickness of the anomalous body. The latter is in accord with minimum depths (25–30 km) estimated from Pn time term. The 6.9-km/sec material is associable with rocks of Mellen gabbro type, and the 6.4-km/sec material with rocks of Duluth gabbro type on the basis of geologic field relations and extrapolated seismic ‘contact outcrops’ supplemented with high pressure and temperature velocity data. The nature of the structural model suggests the 6.9-km/sec material as intruded into the 6.4-km/sec material, which in turn (at least in Wisconsin) is replaced in the upper tens of kilometers with 6.1-km/sec material at a lateral distance of several hundred kilometers from the high. Models across the axial zone of the gravity high in Lake Superior, Wisconsin-Minnesota, and Iowa all show the anomalous high-velocity mass increases in width upward, producing a velocity reversal in the vicinity of the gravity high. A preliminary search for analogous structure among the more modern rift or ridge systems shows that, on the basis of average width, length, compressional velocity, velocity contrast, gravimetric and magnetic expression, structure, etc., the most analogous tectonic feature (at least on the basis of present information) is the Red Sea rift rather than the continental rift valleys of Africa.

Central North American Rift System: 1. Structure of the axial zone from seismic and gravimetric data

Two crustal models, from the Pacific Ocean to the Brazilian shield, have been derived for southern Peru and northern Chile. These crustal sections satisfy both the seismic and major features of the gravity anomaly when using velocity-density relations from Woollard, and Steinhart and Smith. In general, the crustal sections show thick (over 10 km) sedimentary-metamorphic material with velocity about 4.5 km per sec under the altiplano, a 5.7- to 6.1-km-per-sec refractor pinching out near the Peru-Chile trench with a maximum thickness under the altiplano, and a 6.8- to 7.0-km-per-sec refractor becoming deeper and thicker under the Andean altiplano. The crustal thickness is about 12 km under the ocean basin, about 76 km under the altiplano, and 40 km (inferred) under the shield. There is an abrupt lateral structural variability of the two crustal sections that is typical for the area between (at least) 12° and 30° S. lat, as determined from the regional extent of the gravity anomalies. The gravity anomaly map at sea level is dominated by a broad negative anomaly, −400 mgal, easterly offset from the line of highest Andean mountains. The Peru-Chile trench is reflected by a narrow negative anomaly (about −200 mgal extreme) that is separated from the negative Andean anomaly by a narrow, sharp, and relatively positive anomaly occasionally reaching values of about +20 mgal in the vicinity of the shore. The two crustal models account satisfactorily for observed differential time delays between seismic stations near the shoreline and stations on the altiplano. Computations of pressure differential as a function of depth between seismically derived crustal columns under the the Pacific Ocean basin (without removing the water layer) and the Andean altiplano reveal an excess of pressure under the altiplano with respect to the neighboring ocean from the surface to a depth of 55 km, with a maximum (1.5 to 1.7 kb) between 5- and 15-km depth. Below the 55-km depth and down to at least 80 km, the excess of pressure is under the ocean. This differential pressure distribution with depth apparently provides an extra mechanism of stress accumulation influencing the earthquake spatial distribution along the ocean-continent transition zone in western South America.

Leonidas C. Ocola

Papers

Crustal Low-Velocity Zones Under the Peru-Bolivia Altiplano

Central North American Rift System: 1. Structure of the axial zone from seismic and gravimetric data

Crustal Structure from the Pacific Basin to the Brazilian Shield between 12° and 30° South Latitude