Configuración y geoquímica de rocas máficas y ultramáficas de la Formación Cortaderas, suroeste Precordillera
El margen occidental de Precordillera contiene secuencias clásticas distales, de cuenca mas profundas y menores cantidades de calizas del Paleozoico temprano, la procedencia de los cuales era el margen pasivo.
Las rocas máficas y ultramáficas se intercalan en estas rocas sedimentarias de margen pasivo. Los más extensos afloramientos se producen en la zona suroeste como un cinturón de ~ 80 km de largo y 10-15 km de ancho
Nuestro trabajo de campo, geocronológico, y los estudios petrológicos de las rocas máficas y ultramáficas en las áreas de Cortaderas y Pozos delinean tres grandes unidades tectónicas:
Una unidad A formada por gabro, microgabbro, y diabasa (diabasa masiva); un complejo de roca ígnea B fomada por capas ultramáficas y gabros y una unidad C: metasedimentitas con capas de lavas básicas y sills intercalados.
Las dos primeras unidades rara vez se encuentran en contacto tectónico; dónde están, el complejo ultramáfico se superpone a la diabasa masiva; ambos fueron colocadas sobre la tercera unidad a lo largo de zonas de cizalla dúctil de bajo grado, vergentes al este (Fig. 2). El complejo ultramáfico representa un compuesto de rocas cristalinas formadas en diferentes momentos y entornos. Está dominado por las rocas ultramáficas (totalmente serpentinizadas lherzolita, harzburgite y dunite), algunos de los cuales conservan textura de deformación de alta temperatura, y cumulatos ultramáficos (piroxenitas espinela, clinopyroxenitas y hornblenditas). El gabro recristalizó de forma estática a una alta temperatura. El granate es (Davis et al., 2000), localmente abundante y se observa tanto los cumulatos de rocas ultramáficas como en el gneis Qz-FK. La naturaleza del contacto inicial entre las rocas ultramáficas y el gneis es indeterminado porque se sobreimprime por zonas de cizalla de bajo grado. El gabro tiene una firma calco-alcalino con un fuerte agotamiento de las tierras raras livianas (Davis, 1997). Un valor Nd epsilon (eNd) del gabro en capas es 0,6, en contraste con un valor de 7,0 de una lherzolite tectonizada.
Interpretamos esta unidad como una corteza continental profunda underplated por rocas ultramáficas y máficas . Los valores de la geoquímica y eNd de la piroxenita sugieren que no puede ser genéticamente vinculada con el gabro y que puede haber sido más de un evento underplating.
En contraste con el complejo ultramáfica, que se sometió a metamorfismo facies granulita las rocas máficas en las otras dos unidades tectónicas se caracterizan por intrusiones y flujos que se sometieron únicamente a facies de esquistos verdes metamorfismo, prehnite-pumpellyita a nivel superficial.
Las diferencias clave entre la unidad de diabasa masiva y los sills en metasedimentitas son sus edades (se discute a continuación), sus dimensiones y su relación con los metasedimentitas circundantes. La diabasa masiva es de gran tamaño (0,1-0,4 km de ancho, 1-3 km de largo), roca ígnea gruesa de diabasa con características internas tales como cuerpos de plagiogranite en pequeños diques y pillow-lavas. La diabasa masiva está en contacto tectónico con metasiltstone carbonato.
Por el contrario, las coladas de basalto y diabasa en las metasedimentitas son más pequeñas (0,01-0,1 km de ancho y <1 km de largo) y tienen relaciones de deposición e intrusivas con metasedimentitas adyacentes (Fig. 2). Major, oligoelementos, y geoquímica isotópica de las dos unidades tectónicas son indistinguibles, ya que ambos son toleítico con una firma oligoelemento dorsal oceánica enriquecido (E-MORB; Davis, 1997).
Los valores finales de diabasa de ambas unidades están dentro del error de la otra, que van desde 6,0 hasta 7,3. Las firmas geoquímicas y metamórficas de la roca ígnea en la unidad de diabasa masiva son claramente diferentes de la roca ígnea en capas en el ultramáfica capas compleja e indican orígenes diferentes.
Al norte y al oeste, varias unidades contienen características de las rocas metasedimentarias con flujos máficos y soleras. Ordovícico superior flysch de cuenca que contiene los flujos máficas intercaladas y soleras (Fig. 1B) se extiende hacia el norte por casi 5 ° de latitud de las Cortaderas. Estos flujos toleíticos y alféizares tienen E-MORB firmas de elementos traza (Haller y Ramos, 1984;. Kay et al, 1984).
Los dos estudios anteriores también se analizaron las rocas ígneas de Cortaderas y regiones Bonilla y se encontró que tenía una firma de E-MORB, pero las relaciones estructurales de las rocas no se describieron ni eran diferentes unidades tectónicas identificadas. Por lo tanto no podemos correlacionar los datos directamente con cualquiera de las unidades identificadas en este documento. Coladas de basalto y almohadas con E-MORB firmas de elementos traza y el final que van desde 7,1 hasta 8,6 también se producen inmediatamente al oeste de la zona de estudio en el rango de Cerro Redondo, donde son intercaladas con areniscas turbidíticas de edad Silúrico-Devónico (fig. 1B ; Cortés y Kay, 1994)
U/Pb GEOCHRONOLOGY
Samples for U-Pb analysis were collected from
(1) microgabbro within the massive diabase,
(2) layered gabbro within the ultramafic-layered
complex, and (3) a mafic sill hosted in metasedimentary
rocks. The U-Pb analytical techniques
used were outlined in McClelland and Mattinson
(1996). Analytical results are presented in Figure 3
and Table 1.1
Sample 1, microgabbro from the massive diabase,
includes two separate samples collected in
1995 (fractions a–c) and 1997 (fractions d–g)
from precisely the same site and are considered
as a single sample. Five analyses of euhedral lathshaped
zircons (fractions a–e) define a concordia
with upper and lower intercepts of 576 ± 17 and
105 ± 36 Ma, respectively. Subhedral, subequant
grains (fractions f and g) yielded significantly
older 207Pb/206Pb ages of 1031 ± 9 and 1121± 12 Ma and are not on the concordia. The lath shaped
grains are interpreted as xenocryst-free
magmatic zircon formed during gabbro crystallization,
and the subequant grains are inferred to
contain or wholly represent inherited xenocrystic
components. A crystallization age of 576 ±17Ma
is interpreted for the microgabbro.
Samples 2 and 3, layered gabbro of the ultramafic-
layered complex and mafic rock from a sill
in the metasedimentary rocks, respectively, both
yielded one concordant and three discordant
analyses. For sample 2, discordant analyses were
obtained from of subhedral, subequant grains,
and the concordant analysis was obtained from a
single euhedral, flat (aspect ratio 1:2:3) zircon. A
crystallization age of 450 ± 20 Ma is interpreted
for the layered gabbro on the basis of the single
concordant age. For sample 3, there is no obvious
morphological difference between discordant
and concordant fractions. Nevertheless, a crystallization
age of 418 ± 10 Ma is interpreted for the
mafic sill on the basis of the single concordant
age, and the observed discordance is attributed to
inclusion of xenocrystic components. The assigned
uncertainty for both samples reflects the
uncertainty in potential Pb-loss effects as observed
in sample 1, the fact that a singular concordant
analysis does not allow assessment of its
reproducibility, and the fact that concordancy of
these fractions is in part due to the ±10–15 m.y.
uncertainty in the 207Pb/206Pb ages.
DISCUSSION
The spatial association of ultramafic rocks
with gabbro, diabase, and mafic flows led some
authors to refer to the rocks on the western side of
the Precordillera terrane as an ophiolite complex
(Haller and Ramos, 1984; Gregori and Bjerg,
1997), an interpretation incorporated into most
tectonic models to date. This study refutes that interpretation,
because the deep-level ultramafic layered
complex intrudes continental crust and
because the mafic rocks range in age from Precambrian
to Silurian. Rather than describing the
entire assemblage of ultramafic and mafic rocks
as an ophiolite, we limit application of this term
to the massive diabase. Regardless, the interpretation
by Ramos et al. (1986) that these ultramafic
and mafic rocks mark a suture between two continental
terranes, the Precordillera and Chilenia,
has not changed significantly. We present the following
interpretation of the tectonic setting for
each of the three tectonic units, using an assumption
that the structurally highest and farthest west
unit, the ultramafic-layered complex, formed at
the greatest distance from the Precordillera margin.
Detailed discussion of the structural setting of the rocks of the southwest Precordillera is beyond the scope of this paper and can be found in Davis et al. (2000). Several aspects of the massive diabase unit argue that it formed in a well-developed rift or transition to early drift phase of separation of the Precordillera from Laurentia, including thickness and homogeneity of the unit, sheeted dikes with pillow screens, coarse gabbro and plagiogranite bodies, lack of interlayered sedimentary and silicic volcanic rocks, and moderately depleted Nd isotope signature, all of which allow it to be described as an ophiolite.
The main arguments against this intepretation are that the sedimentary record of the Precordillera puts the maximum age of the rift-drift transition in the Early Cambrian (544–511 Ma) (Thomas and Astini, 1996). This time coincides with the oldest igneous rocks erupted in the Ouachita embayment, which led Thomas and Astini (1996) to propose this site as the origin for the Precordillera terrane. The age of 576 ± 17 Ma for the massive diabase, however, correlates more closely with the youngest synrifting volcanic rocks from the Blue Ridge rift (564± 9 Ma) (Aleinikoff et al., 1995), the Late Proterozoic Laurentian margin north of the Ouachita embayment. Thus our model (Fig. 4) shows the massive diabase forming in a widening intracontinental rift substantially earlier than the Early Cambrian age proposed by Thomas and Astini (1996) for initiation of rifting on the western margin of the Precordillera.
The other displaced tectonic unit, the ultramafic-
layered complex and associated quartzofeldspathic
gneiss, raises the critical question of
whether the layered gabbro intrudes Precordillera
basement or some other continental crustal mass,
the enigmatic Chilenia. The most compelling evidence
that Chilenia is distinct from the Precordillera
is in the pronounced change in isotopic
and trace element geochemistry of Neogene volcanic
rocks west of Cerro Redondo (Fig. 1B)
(Kay and Abbruzzi, 1996). If the layered gabbro
intrudes Precordillera basement, then it would
have formed in a rift setting on the western margin,
as this region underwent extension in the
Middle and Late Ordovician (Kay et al., 1984;
Keller et al., 1998).
An alternative is that the layered gabbro formed during initiation of convergence along a plate margin distant to the Precordillera, during plate reorganization following collision with the Famatinian arc (Astini et al., 1995; Davis et al., 2000).
The preliminary Nd isotope data reported here from the layered gabbro have a distinctly more enriched mantle signature than any of the other igneous rocks. The calcalkalic character of the layered gabbro and the spatial association with pargasitic hornblendite dikes support the intepretation that the layered gabbro complex, and at least some of the ultramafic rocks, formed in the hanging wall of a subduction zone. Thus the chemical data and its structural position, at the top of the tectonic sequence, argue strongly for the continental crust of the ultramafic-layered complex being derived from a terrane distinct from the Precordillera, perhaps Chilenia (Fig. 4).
The Silurian sills, and lavas to the west in distal
sedimentary units (Cortés and Kay, 1994),
suggest that the tensional history of the western
margin continued intermittently, following Ordovician
extension. The striking chemical similarity
of the Silurian sills to the Precambrian diabase
and Ordovician pillow basalts farther
north supports the idea that all formed from the
same part of the mantle on the western margin of
the Precordillera crust. Several possible scenarios
could explain the formation of the sills and
flows (Kay et al., 1984); a transform-dominated
rift setting (Fig. 4) is one environment compatible
with a history of convergence and sinistral
strike-slip deformation continuing on the eastern
side of the Precordillera and western Sierras
Pampeanas (Ramos et al., 1998; Rapela et al.,
1998), synchronous with the sills intruding on
the western side of the Precordillera.
Assembly of the structural section currently
observed in the western Precordillera occurred
primarily in the Early to Middle Devonian (Davis
et al., 2000). This convergence coincided with
formation of the youngest dates on shear zones in
the Sierra Pampeanas (Ramos et al., 1998) and
brought to a close the series of arc-continent collisions
along this part of the Gondwana margin.
CONCLUSION
New U-Pb zircon ages and initial Nd isotope
data from mafic rocks on the western margin of
the Precordillera, in conjunction with published
Ar/Ar dates from metamorphic rocks, indicate
that a long history of tectonism is preserved in a
very narrow zone. Precambrian, Ordovician, and
Silurian extension-related shallow-level mafic
intrusive and volcanic rocks on the western margin
of the Precordillera terrane were imbricated
in a Devonian collision that emplaced part of a
Late Ordovician arc and associated continental
crust onto the western margin. The inception of
the Late Ordovician arc coincided with extension
on the western margin of the Precordillera,
indicating a time of major plate reorganization following collision of the Precordillera with the
Famatinian arc to the east. Thus the new data
support a model of development of ocean crust
west of the Precordillera prior to the Late Ordovician,
as proposed by Astini et al. (1995) and
Thomas and Astini (1996).
GEOLOGIA DEL COMPLEJO BONILLA
East of Uspallata village, appears a 30 km N-S long and 8 km wide belt of slightly metamorphosed rocks assigned by Keidel (1939) to “Conjunto de Farallones, Conjunto de Bonilla, Manto de Buitre, and Manto de Jagüelito” . Later, the formers three were assigned to formations (Borrello, 1969; Varela, 1973) and included in the Bonilla Group by Folguera et al (2004).
The Manto de Buitre was considered as Devonian (386 ± 20 and 369 ± 20 Ma) by Cortes et al., (1997) and Folguera et al (2004), due to a radiometric dating of an alkaline mafic body emplaced in this unit at Quebrada de los Burros.
According to the above cited authors all units consists of quartz schists, phyllites, greenschists, marbles, limestones, dolomites, meta-gabbros, and serpentinites. Nearly all units are partially cut by mafic dikes and sills interbedded in the sequences. Due to the heterogeneous composition of this succession that includes sedimentary, metasedimentary and mafic and ultramafic rocks, we propose to name this unit as Bonilla Complex. Accordingly, we will describe minor units as lithofacies.
The maps and detailed profiles allow to recognize five lithofacies of the Bonilla Complex as follows:
- Facies of sandstones, calcareous sandstones, pelites, marls, limestones and small mafic bodies.
- Facies of graphitic phyllites, slates, phyllites with intercalations of feldspar-rich quartzites and marls.
- Facies of laminated and crystalline limestones and marls.
- Facies of mafic bodies.
- Facies of ultramafic bodies.
I Sandstones, calcareous sandstones, pelites, marls, limestones and small mafic bodies
These rocks crop out in the western part of the belt, between Quebrada del Guanaco Muerto and Quebrada Angosta (Figs. 2, 3, 4 and 5) forming the western slope of the Cordón Jarillal, as well as, the summit and the western slope of the Cordon de Bonilla.
This facies is equivalent, in part, to the Farallones Formation. The metasedimentary pile is located above the Carboniferous marine Santa Elena Formation due to a back-thrust that mark the western extension of the Bonilla Complex. Since this facies is slightly folded and shows a nearly constant dip of around 25 degrees to the east, forming a homoclinal structure, we have been able to establish its thickness, which reaches 1,000 m.
As depicted in figure 7b sandy levels, which are not the most abundant, pass transitionally to calcareous sandstone bars. These structures appear countless times in the sequence. Individual thicknesses range from 1 to 2 m. Sandstones are composed mainly (75 %) by rounded to subangular grains of quartz, detrital micas, and variable quantities of calcite. Groundmass is composed by pale color fine grained micas and chlorite, located between quartz grains or forming small ribbons along stratification (Fig. 7g).
The lithological characteristics and the metasedimentary structures show that facies I can be considered to represent the shallow portion of a marine platform, possibly near the coast, strongly affected by the action of tides and waves. The measurement of festoon structures in sandstones, as well as, the architecture of the calcareous sandstone bars indicates a flow from the east, considering actual coordinates.
II Marls, black phyllites and phyllites
This facies forms two domains, one located in the central part of the belt and the other in the eastern border, both more than 1 km wide in W-E direction (Fig. 3, 4, 5 and 7a). The central domain is transitional to facies I in the eastern slope of Cordón Jarillal, whereas the eastern one is unconformably covered by Permian volcanic rocks of the Choiyoi Group. The central extends from the Quebrada Guanaco Muerto to the Mogote de los Gateados, (Fig. 3) in the western slope of the Cordón de Bonilla.
The eastern domain extends between Cerro El Sapo and Cerro Jagüelito, disappearing southwards. In the type area, located east of Cordon del Jarillal, shows a thickness near to 1,000 m. This thickness is tentative because in this area several ultramafic bodies were tectonically emplaced, producing deformation in the host rocks.
The most abundant rocks are laminated marls. They are east-dipping and show scarce deformation, excepted in the vicinity of the ultramafic bodies, where local deformation is important. Texturally, they show a micrometric lamination, mostly composed by bands of small mica flakes, elongated within the stratification. Abundant, but isolated spots, 500 x 250 μm of calcite appear irregularly scattered between the micas. Some of them are aligned with the stratification and form short ribbons.
The phyllites are generally gray color and are interbedded within the marls, also showing a millimetric lamination. These rocks, as indicated above, show an alternated laminated stratification, 1 to 2 mm thick. Bands are of quartz, mica and carbonate. The first appears as small, round grains, showing sometimes ondulose extinction. Size average is 120 μm, but bigger, subhedral, crystals, possibly related to incipient recrystallization can be observed. Micas appear in ribbons, 1 mm thick, with flakes growing parallel to the stratification. Sericite is common, but small, scarce, detritial muscovite occurs occasionally. Chlorite was also observed. Micas show bending, kink bands and are refolded.
The black phyllites and slates are located mainly near the contact with facies I, but occasionally appear in the Cerro Jagüelito area. They show a micrometric lamination, composed mainly by a fine grained dark material, possibly organic matter, and small quantities of quartz and calcite. All mineral are aligned along the stratification.
Sporadically, in the central domain of facies II are recognized levels of limestones, which become more common eastwards, being transitional to facies III.
According to the abundance of fine grained sediments, absence of high-energy sedimentary structures and their homogeneous lithology we considered this facies as representing the middle to lower part of an offshore platform, away from the action of waves and tides.
III Laminated and crystalline limestones and marls
This facies crops out in the eastern part of the belt, forming the southern tip of the curved Sierra de Uspallata (Fig. 5). It appears in the Cerro El Sapo area, along the Sierra de Uspallata until the Rivadavia Mine. Eastwards is transitional to the eastern domain of facies II. Southwards, there are small spots north of Los Chuecos Mine, but near Cerro Jaguelito and Portillo de Bonilla bigger outcrops appear. South of Quebrada de los Burros there are few outcrops of these rocks. The thickness of this facies can be estimated as near to 500 m in the area of Cerro El Sapo, increasing to 1 km east of the Rivadavia Mine.
A good section of this facies can be observed immediately east of the Rivadavia Mine. There, the section starts with interbedding between marls of facies II and limestones of facies III. Small levels of compact pale orange limestones (Fig. 7c) occur in between the eroded light gray marls, which further east, are replaced with abundant laminated limestones. In this area, the limestones are between 1 and 30 cm thick.
They were west-vergent folded and refolded in a centimetric to metric scale). The folding is strongly asymmetrical, with high-angle limbs dipping to the west and gently to the east. In many cases recumbent folds are recognized in the western part of the outcrops. Small-scale back-thrust displacing the limestones and emplacing little fragments of mafic rocks in the sequence can be observed along the profile.
Reduced slices of marls and phyllites were occasionally recognized, passing quickly to the laminated limestones. Limestones are fine-grained, mostly composed by sub-rounded grains of calcite and a few of dolomite (Fig. 7e). Crystalline limestones are rare and reduced to small levels associated to movements between strata.
IV Mafic bodies
The mafic bodies are dikes, sills and layers of different sizes interbedded and emplaced in the metasedimentary sequence. Dikes are common in the Quebrada Guanaco Muerto area. They are subvertical, up to 7 m thick and cut sharply some sections of the facies I.
In the Mendocina Mine area form layers that follows the stratification of the metasedimentary rocks. Sizes vary between 10 cm to 20 m thick, with longitude between 5 and 100 m. Near the Mendocina Mine occur a 40 m thick, 150 m log body interbedded in the metasedimentary sequence of facies I. This body shows a porphyritic texture in the contact with the sediments, coarsening to the centre of the body to an inequigranular texture. The borders of these bodies show a cleavage that mimetic those of the metasedimentary host rock. Layers of mafic bodies are relatively common in the facies I, west of the Mendocina Mine area.
Sills are rare and appear in the eastern part of the Quebrada Guanaco Muerto area, westward of Agua de la Cueva and in the Quebrada de los Burros (Fig. 7f).
These bodies are nearly elliptical, with its long axis exceeding 150 m and minor about 40 m. Most are emplaced in facies II and a few in facies I. They show sharp contact with the host rocks, cutting the stratification of the metasedimentary sequences, and seem to represent the pipes that fed the largest subvolcanic bodies, although such association was never recognized in the field. They have an intergranular texture composed by plagioclase laths, (70 %), biotite (10 %), pyroxene (5 %), apatite, illmenite and opaque minerals. Large plagioclases (1 mm long) enclose interstitial crystals of pyroxene, biotite and other minerals. There are not evidences of deformation, but alteration of plagioclases to calcite and sericite is common.
V Ultramafic bodies
These rocks appears as isolated major, or as strings of minor bodies tectonically emplaced in facies II and less in facies I. Those emplaced in facies II are located northwards of the Rivadavia and Mendocina mines. An ultramafic body located 3 km NW of the Cerro El Sapo is 100 m thick, 1.3 km long and 800 m wide and is overthrusted top-to-the-west on the phyllites of facies II. Another isolated body (85 m x 250 m x 350 m) is located 1 km NE of Cerro El Sapo, but its contact with rocks of facies II is unclear.
A string of N-S elongated bodies can be observed between Quebrada Guanaco Muerto and Quebrada de Santa Elena, along 6 km. Between Mendocina Mine and Quebrada Guanaco Muerto appear four ultramafic bodies, being the most important one that is 1.5 km long and 400 m wide. In the Quebrada Guanaco Muerto a small, oval body, shows intense foliated borders and mineral lineation indicative of top-to-the-west movement. From Mendocina Mine to the south the string is formed by smaller bodies that follow a N340° strike and are emplaced along a strike-slip fault that intersect the Cordón de Bonilla (Fig. 4). Further south the ultramafic bodies disappear.
Another string of ultramafic bodies are located 500 m eastwards to the above described. Here, bodies are more elongated (200 m long, 80 m wide) and looks more affected by the later deformation. The body at Rivadavia Mine is the biggest of this area, being 400 m wide. The eastern border of this string is marked by a top-to-the west overthrust of the phyllites of facies II on the ultramafic bodies. Small ultramafic bodies, possibly belonging to this string can be observed few kilometers west of Cerro Jagüelito. These rocks show a compact texture, composed mainly by serpentine mineral forming a mesh of flakes with different orientations. Chlorite and talc are also abundant. Relicts of olivine and pyroxenes can be recognized (Fig. 7h).
Geochemistry of the Bonilla Complex
Geochemical investigation has been carried out by analyzing 17 samples for major and trace elements of the metasedimentary rocks, 13 of the mafic rocks and 10 of the ultramafic rocks. Results are listed in table 1, 2 and 3 respectively. The analyses were made using INAA and ICP-MS at ACTLABS (Canada).
Geochemistry of the metasedimentary rocks
Rocks classified in the field and microscopically as sandstones, pelites and limestones plot in the Al2O3+Fe2O3-SiO2-MgO+CaO diagram of Pettijohn et al. (1987) in the sandstones, limestone and lutite-pelite fields respectively. If the Na2O-Fe2O3t+MgO-K2O triangle of Blatt et al. (1980) is considered, samples plot in the lithic sandstone and greywacke fields. The log SiO2/Al2O3-log Na2O/K2O diagram of Pettijohn et al. (1987) shows greywacke and lithic arenite compositions.
Diagrams and chemical analyses indicate that these lithologies show a relatively restricted range in major and trace elements chemistry. Silica varies between 67 and 76 wt%, whereas alumina between 9 and 13 wt% and Na2O between 0 and 6 wt%.
Provenance
The Zr-15*Al2O3-300*TiO2 diagram of Garcia et al. (1994) indicates that greywacke and lithic arenite conform one trend extending from feldspar-rich sandstones towards mature sandstones, mostly concentrated in immature sandstones (Fig. 8d). This evolution trend is concordant with the Al2O3/SiO2 ratios, which change gradually from 0.20 to 0.12, with most samples concentrated between 0.16 and 0.14, which reflects a high degree of sorting during sediment transport.
The relative abundance of Al2O3, CaO, Na2O and K2O as shown on the molar-proportion diagrams (Nesbitt and Young, 1996) are consistent with typical continental alteration trends influenced by weathering of plagioclase-alkaline feldspar to form illite-sericite (Fig. 8e). No evidence of amphiboles or pyroxenes in the original sediments can be recognized through these diagrams. Altogether, as well as, the low concentration of MgO, Fe2O3 and TiO2 suggests no contribution from mafic source rocks. Metasedimentary rocks of the Cordón del Portillo (Vujovich and Gregori, 2002) and the Guarguaraz Complex follow that trends.
The trace element data were plotted on a Th/Sc vs. Cr/Th diagram (Totten et al., 2000). This plot contrasts the different proportions of continental and mafic sources. As shown in Figure 8f, samples of the Bonilla Complex plot near the average of upper continental crust, which is indeed near the average of granitic rocks, discarding a two-component mixing model between felsic and mafic end members. Samples are nearly coincident with those of the Guarguaraz Complex and the metasedimentary rocks of the Cordón del Portillo. The La/Sc vs. Th/Sc diagram (Totten et al., 2000) indicates that the rocks of the Bonilla Complex plots around the field of NASC (North American Shale Composite, Haskin and Haskin, 1966), being equivalent with those of the Guarguaraz Complex and the metasedimentary facies of Portillo area (Vujovich and Gregori, 2002)
All rocks, including the Metasedimentary rocks of the Cordón del Portillo (Vujovich and Gregori, 2002) and the Guarguaraz Complex display a positive correlation between K and Rb, (Fig. 8h), which is similar to that of the upper crustal rocks (Taylor and McLennan, 1985).
Major element analyses
Roser and Korsch (1986) have developed a bivariant tectonic discriminator for sandstones and mudstones using SiO2/Al2O3 vs. K2O/Na2O. The fields are based on ancient sandstone-mudstone pairs, cross-checked against modern sediments from known tectonic settings, from which they differentiated sediments derived from volcanic island arcs (ARC), active continental margin (ACM) and passive margin (PM). Roser and Korsch (1986) tectonic diagram indicates that the samples from the Bonilla Complex are PM-derived, which imply that metasedimentary protoliths derived from stable continental areas and were deposited in intra-cratonic basins or passive continental margins (Fig. 9a). Similar results were obtained for the Guarguaraz Complex by López and Gregori (2004).
The discriminant functions of Roser and Korsch (1988), using Al2O3, TiO2, Fe2O3t, MgO, CaO, Na2O and K2O contents as the variables, were designed to discriminate between four sedimentary provenance types: mafic (P1, ocean island arc source, similar to the ARC-derived), intermediate (P2, mature island arc, also similar to ARC-derived), felsic (P3, ACM-derived), and recycled (P4, granitic-gneissic or sedimentary source area, similar to PM-derived).
The majority of the metasedimentary rocks of the Bonilla Complex, as well as the Cordón del Portillo metasedimentary rocks, plot within the P4 (recycled) and P2 fields (Fig. 9b), supporting the interpretation that they are derived from a craton interior or a recycled orogenic terrane.
Normalized multi-elements plots
Trace-elements plots are useful for tectonic setting discrimination, although some diagrams are influenced by sorting, heavy mineral contents and proportion of mafic input.
Floyd et al. (1991) proposed to use the full range of elements for tectonic setting discrimination, in order to avoid spreading of rock series across geologically unrelated tectonic fields. Upper continental crust-normalized multi-element plots of Taylor and McLennan (1985) were used by Floyd et al. (1991) to compare the compositions of greywackes and sandstones from different tectonic environments: passive margin (PM), oceanic island arc (OIA) and continental arc + active margin (CAAM). The elements are arranged (from left to right) in order of decreasing ocean residence times, and comprise a mobile group (K-Ni) and a relatively stable group (Ta-Th). Diagrams of the Bonilla Complex (Fig. 9c, d and e) were compared with average greywackes from PM, CAAM and OIA settings, being our interpretation as deposited in a passive margin environment.
Rare Earth elements analyses
Rare earth elements are considered to be immobile under weathering, diagenesis and up to moderate levels of metamorphism and, for clastic sediments, its concentrations is equivalent to those of the source rocks.
In the REE spidergram (Fig. 9f), rocks of the metasedimentary assemblage show similar patterns to PAAS (Post-Archean Average Australian Sedimentary rocks, McLennan, 1989) and NASC (North American Shale Composite, Haskin and Haskin, 1966). They are between 80-200 times LREE enriched, with a nearly flat HREE pattern 10-30 times enriched, and small negative Eu anomalies. They are nearly coincident with the metasedimentary rocks of the Guarguaraz Complex and the Portillo area (Vujovich and Gregori, 2002).
Geochemical signature of REE suggests that the rocks of the Bonilla Complex are composed of sediments derived from old upper continental crust and/or young differentiated arc material. According to McLennan et al. (1990), these provenance components are found in several basin types, but rarely in a fore-arc setting.
Geochemistry of the mafic rocks
Major element analyses
Metabasites from the Bonilla Complex have SiO2: 41- 51 wt %, Al2O3: 12-18 wt %, and MgO: 3-9 wt %, classifying as picrobasalts, basalts, basaltic andesites and basanites in the TAS diagrams (Fig. 10a) of Le Maitre et al., (1989) showing a more restricted composition when compared with samples of the Guarguaraz Complex.
The TAS vs SiO2 diagram (Fig. 10b) indicates an alkaline composition, overlapping in part the composition of the Guarguaraz Complex and Metales belt metabasites (Gregori and Bjerg, 1997). A tholeiitic behavior can be recognized in the AFM diagram of Irvine and Baragar (1971) follow a similar but more restricted trend that the samples of the Metales belt metabasites (Fig. 10c).
In the Ni-MgO diagram samples of the Bonilla Complex lay on the field of the Guarguaraz Complex (Fig. 10d) and also over the dikes of the Troodos complex and the E-MORB basalts of the Lau basin. In a FeOt/MgO-TiO2 diagram (Fig. 10e) the mafic rocks assemblage exhibits few correlation with the Reykejanes Ridge basalts (Tarney et al., 1979) and other rocks used for comparison, possibly due to alteration.
Trace elements
Trace elements concentrations were normalized to EMORB (Fig. 10f), according to Sun and McDonough (1989). Most samples have a very similar pattern for elements ranging from Sr to Yb, displaying EMORB signatures. All samples show Pb anomalies of 8-20 times over EMORB, indicating crustal (sediment) contamination.
Rb, Nb and K are scattered, possibly due to mobility in aqueous phase (Pearce, 1992).
The LILE have the following maximum concentrations: Ba (703 ppm), Sr (920 ppm), Rb (68 ppm). Values for HFSE are: Ta (4.9 ppm), Nb (63 ppm), Hf (8 ppm) and Th (6 ppm). Cs shows a higher concentration than EMORB. In the Th/Yb versus Ta/Yb diagram (Pearce, 1983), which differentiate between subduction-related basalts and oceanic basalts, samples display compositions ranging between MORB and EMORB (Fig. 10g) with mixed signatures of depleted and enriched mantle source, excluding arc signatures.
Rare Earth elements
Most samples show low concentration in REE (180 ppm). They are 10 to 100 times LREE enriched with small positive Eu anomalies, due to plagioclase fractionation (Fig. 10h). Most samples can be classified as EMORB. Samples of the Bonilla Complex have a similar composition to the Precordilleran Ordovician basalts (Kay et al, 1984) and the Guarguaraz Complex (López and Gregori, 2004).
Geochemistry of the ultramafic rocks
Major elements
Ultramafic rocks have SiO2: 38-41 wt %. They plot near the FeOt component of the CaO-FeOt-Al2O3 system, indicative of the absence of strong alteration in these rocks. Also they plot away of the ultramafic rocks of the Guarguaraz Complex, which are strongly affected by deformation and alteration, due to the presence of intrusive acidic bodies (Fig. 11a). In the Px-Pl-Ol diagram (Yoder and Tilley 1962) they classify as Pl-bearing ultramafic rocks, (Fig. 11b), although its quantities are very small, and can not be compared with the samples of the Guarguaraz Complex, which are considerably more rich in plagioclase. This result is also corroborated in the CaO-Al2O3-MgO diagram of figure 11c, in which samples plot near the MgO component, indicative of minor quantities of plagioclase in the rock. In the Opx-Cpx-Ol diagram for ultramafic rocks (Fig. 11d), samples plot mainly as olivine-websterites and lherzolites. The ALK-FeOt-MgO diagram of figure 11e shows that near all samples fall in the MgO rich component in a composition similar to those of the mafic and ultramafic cumulates of the ophiolitic sequences.
Compared with average komatites, mafic and ultramafic cumulates, the ultramafic rocks of the Bonilla Complex can be assigned to ultramafic cumulates and lherzolites (Fig. 10d).
Rare Earth elements
Maximum REE concentration for the ultramafic rocks is 154 ppm, but this result together with that of sample 7090292U seems to be anomalous, because most samples have concentrations between 2 and 7 ppm.
In a chondrite-normalized REE plot (Fig. 11f), samples with the highest REE concentration show 100 times enrichment over chondrite in LREE. These rocks display a pattern more characteristic of a plagioclase-rich gabbro. The second group shows nearly flat patterns slightly enriched over chondrite. Some have a small negative Eu anomaly and other some irregularities in Tb, Dy and Ho concentrations, possibly due to hydrothermal alteration. These rocks are similar to samples of the Ussuit Komatiite (Kalsbeek and Manatschal, 1999).
Detrital zircon U–Pb geochronology of the Bonilla Complex
Two detrital zircon samples from the Bonilla Group were analyzed for geochronology.
Both belongs to facies I. Sample B1 is located at S 32°35´56´´- W 69°13´47´´ in the Quebrada Santa Elena and correspond to a quartz–mica sandstone (Fig 6, profile a). A total of 98 zircons from sample B1 were analyzed. A large part (Table 4) of the zircons formed during an age interval of 0.7–0.5 Ga. (Fig. 12a). Other, less pronounced age accumulations can be found at 1.5–1.0 Ga. There are single ages at 2.4 Ga and 2.8 Ga.
Sample X4 is a sandstone of the lower part of the sequence located west of Portillo de Bonilla at 32° 39´13´´ S-69°13´14´´W and also correspond to a quartz–mica metasandstone of facies I. Zircons from sample X4 provided 96 analyses, which define six dominant ages of 592, 656 Ma, 713 Ma, 1135 Ma, 1242 Ma and 1950 Ma. Two grains give ages of ca. 2.6 and 2.8 Ga (Fig. 12b). Maximum age for B1 sample is 510 Ma whereas for X4 sample is 592 Ma, which indicate Ediacaran to the boundary between series 2 and 3 of Cambrian times (International Commission on Stratigraphy, 2009).
Rapela et al. (1998) report U–Pb SHRIMP detrital zircon ages from a high-grade metapelitic migmatite in the Rio del Suquía area of the Sierras de Córdoba, with ages at 600–650 and 800–1000 Ma and two grains older than 1400 Ma.
In the Sierras Pampeanas of San Luis, Sims et al. (1998) report U–Pb SHRIMP analyses of detrital zircon grains from four high-grade metasedimentary rocks, which show clusters of ages ranging between ~500–700 and 900–1100 Ma, and more isolated results between 1.45 and 2.5 Ga.
Schwartz and Gromet (2004) analyzed a sand/silt layer within pelitic gneisses located near La Puerta, and a quartz-rich layer in the Tuclame Formation of the Sierras de Córdoba. Of the 19 grains analyzed, 5 yielded ages between 550 and 750 Ma, 2 around 850 Ma, 11 yielded ages between 950 and 1050 Ma, and 1 grain shows an age of approximately 1900 Ma.
In the Quebrada de La Cébila, southwestern Sierra de Ambato, Catamarca province, (Fig. 12c), Rapela et al (2007) sampled quartz-feldspathic metasedimentary rocks of La Cébila Formation. The age pattern is dominated by Neoproterozoic zircons with a peak at 640 Ma, and a subpopulation at ∼530 Ma. Minor peaks occur at 790 Ma, 1.77 and 2.1 Ga with minor contributions of Mesoproterozoic and Archean ages.
In the Sierras Pampeanas de Catamarca, the Ancasti Formation (Willner et al., 1983) is composed by banded schists. Rapela et al (2007) report ages of detrital zircons of this unit. They have a bimodal age distribution, dominated by Mesoproterozoic (1100–960 Ma) and Neoproterozoic (680–570 Ma) components, with youngest grains indicating a Neoproterozoic depositional age (≤570 Ma). There are minor component represented by Late Paleoproterozoic concordant grains (∼1.85 Ga), with absence of concordant zircons in the 2.02–2.26 Ga interval (Fig. 12d).
In northern Patagonia, the El Jagüelito and Nahuel Niyeu formations have Cambrian trace fossil content typical of the Sierras Pampeanas of northern and central Argentina (González et al. 2002). Zircon ages from the El Jaguelito Formation (Fig. 12e) have their main provenance at 535–540 Ma suggesting Early Cambrian deposition (Pankhurst et al., 2006). Such a pattern of age distribution is nearly coincident with sample B1 of the Bonilla Complex.
The Nahuel Niyeu Formation (Fig. 12f) display zircon age at ~515 Ma and ~1.0 Ga, as well as, a small component of older ages, including some at ~2.2 Ga (Pankhurst et al., 2006). It is comparable with sample X4 of the Bonilla Complex.
In the Cordillera Frontal (Willner et al., 2008), the detrital zircons of the metamorphic rocks known as Guarguaraz Complex yield ages of 555, 581, 935, 1092, 1228, 1361, 1794, 1893, 2505 and 2788 Ma. This sample show coincidences with sample X4 of the Bonilla Complex which also display ages of 592, 1135 Ma, 1242, 1950, 2600 and 2800 Ma (Fig. 12g).
The Puncoviscana Formation of northwestern Argentina (Fig 12h) show important contribution at 523, 534, 551, 583, 612, 623, 650, 979, 1000, and minor at 2370, 2513 and 2590 Ma. (Adams et al, 2010).
The bimodal pattern of the Bonilla Complex, with peaks at 1.5–1.0 Ga and 0.7-0.5 Ga (Fig. 12), is coincident with the pattern of the Puncoviscana Formation and also with the metasedimentary rocks of the Sierras Pampeanas of Catamarca, Córdoba and the Sierra de San Luis (Sims et al., 1998; Pankhurst et al., 2000; Schwartz and Gromet, 2004; Escayola et al., 2007 Steenken et al., 2006, Adams et al, 2010), indicating that the source area for the whole Pampean belt and the Bonilla Complex was dominated by Meso- and Neoproterozoic components. The presence of ∼1.8 Ga Late Paleoproterozoic zircon in the B1 sample and ~1.9 Ga zircon in the X4 sample has been also observed in the Sierras de Córdoba (Schwartz and Gromet, 2004) and in the Sierra de Ancasti (Rapela et al., 2007).
According to Pankhurst et al., (2006) the low degree metamorphic rocks of the northeastern North Patagonian Massif appear to have been originally deposited as sediments on a continental shelf at the southern margin of Gondwana, underlain by basement rocks that were already part of the Gondwana continent by Cambrian times.
All units show important contributions around 500 Ma and 1.0 Ga, with peaks at 2.2, 2.3 and 2.65 Ga. Comparison of samples B1 and X4 of the Bonilla Complex with the zircon patterns of the Cambrian low-grade metasediments of NE Patagonia (Pankhurst et al., 2006), (the Nahuel Niyeu Formation e.g.), is straightforward.
Discussion
Provenance and correlation
Petrographic and geochemical studies reveal that the protoliths of the Bonilla Complex were dominantly quartzitic and feldspathic in nature. Architectural sedimentary arrangements indicate that these rocks were deposited in internal and external platform environments, whereas measurements of cross-stratification and festoons show paleocurrents from the northeast and southeast (actual coordinates).
The geochemical signature of the protolith, which is typical of felsic to intermediate upper continental crustal rocks, suggests the presence of an older exhumed basement eastwards (actual coordinates) of western Precordillera. Candidates are the actual Sierras Pampeanas of Córdoba, San Luis and San Juan.
Extensional setting for the deposition of the Bonilla Complex is also recorded by the geochemical signatures of interbedded mafic sills and dikes.
Metasedimentary rocks with similar characteristics can be recognized in western Precordillera (Cortaderas area, Davis et al., 2000). There, in the marine sequences, were emplaced microgabbros (576 ± 17 Ma), gabbros and ultramafic rocks (450 ± 20 Ma) and mafic flows and sills (418 ± 10Ma).
In the Cordillera Frontal (Fig. 1), the Guarguaraz Complex is composed of calcareous metasiltstones, metasandstones, schists, phyllites and marbles, interbedded with metadiabases and ultramafic bodies. Dessanti and Caminos (1967), Caminos (1993) and López and Gregori (2004) considered the Bonilla Complex as being equivalent to the Guarguaraz Complex.
The correlation of the Bonilla Complex with the Guarguaraz Complex and rocks in the Cortaderas area suggests that these units were accumulated in an Ediacaran to Middle Cambrian passive margin (Fig. 12c). The maximum depositional age of the Bonilla Complex (592 Ma), the Guarguaraz Complex (556 Ma) and the Cortaderas Formation (early to 576 ± 17 Ma, see above) are coherent with the existence of such an open marine environment in the western (actual coordinates) margin of Gondwana. The presence of Bavlinella and Leiosphaeridia in the Guarguaraz Complex (López, 2005), comparable with palynomorphs of the Corumbá Group (Misi et al., 2006), supports an Ediacaran age.
Therefore, an open marine basin was developed in the western border of the Gondwana margin considerably earlier (~ 50 Ma) than the supposed detachment (~524-~510 Ma, Thomas et al., 2001) of the Cuyania Terrane from the Ouachita embayment in the Laurentia margin. Indeed, these ages are consistent with the younger pulse of volcanic (570-560 Ma) and intrusive activity (680 ± 4 Ma-562 ± 5 Ma), that led to continental separation and the opening of the Iapetus Ocean (Aleinikoff et al., 1995; Tollo et al., 2004).
Gondwanan source of ~500-600 Ma zircons.
In the Bonilla Complex, zircons population formed in between 0.7 and 0.5 Ga is coincident with the Brazilian or Pampean tectonic event. Prominent peaks are at 510, 592, 611, 656, 713 and 720 Ma. Secondary can be distinguished at 1012-1220 Ma, 1135 Ma, and 1242 Ma.
According to Schwartz and Gromet (2004), the detrital zircons from the metasedimentary assemblages of the Sierras de Córdoba display age distributions which point to typical Gondwana contributions (Brazilian: 500–700 Ma; Late Mesoproterozoic ~1000 Ma) and Transamazonian 1900–2200 Ma) provenances.
The Neoproterozoic Brazilian–Pan-African orogeny was a major amalgamation event in the development of Western Gondwana, and Neoproterozoic plutonic bodies are widespread in the Atlantic Shield region of eastern South America. The Brazilian belts, dated 0.53 to 0.93 Ga, are prevalent over the Brazilian Shield and resulted from the convergence of the Amazonia, the São Francisco craton and other crustal blocks during the final assembly of Gondwana (e.g. Pimentel et al., 1999; Brito-Neves et al., 1999; Kröner and Cordani, 2003; Veevers, 2004; Vaughan and Pankhurst, 2008; Cordani et al., 2009).
Thus, both the magmatic activity and the exhumational/erosional history of these regions support the idea that the São Françisco, Borborema and Mantiqueira provinces provided detritus during the development of marginal basins to South America for the Late Neoproterozoic to Early Cambrian period.
Further west in central Brazil, the western Goiás massif in the Tocantins Province hosts another important collection of intrusive granitic plutons of Late Neoproterozoic–Early Cambrian age.
The presence of Late Neoproterozoic plutonic bodies undergoing uplift and denudation between 580 and 460 Ma fits remarkably well with detrital zircon data from sediments in the Puncoviscana Formation, the rocks of the Sierras Pampeanas, the Bonilla Complex and the low-degree metamorphic rocks of northern Patagonia.
Furthermore, Chew et al. (2008) proposed that a Neoproterozoic (Brazilian) active margin belt is buried under the present-day Andean Belt, according to the U–Pb ages of detrital and inherited zircons from sedimentary and granitoid rocks of the central to northern Andean area.
Late Neoproterozoic postorogenic Brazilian belts such as the Tucavaca belt south of the Sunsás belt and the Araguia–Paraguay belt (both 0.5- 0.6 Ga; Litherland and Bloomfield, 1981; Pimentel et al., 1999) as well as the Sierras Pampeanas basins in northern Argentina (0.52- 0.56 Ga; Rapela et al., 1998a) were probably formed during the rifting of eastern Laurentia from southwestern Gondwana and the opening of the Iapetus Ocean (Grunow et al., 1996; Trompette, 2000; Ramos, 2008; Cordani et al., 2009).
The Mesoproterozoic detrital zircons would have been from the southwestern margin of the Amazonian craton, which is marked by three belts: the Sunsás belt, the Aguapeí belt and the Nova Brasilândia belt of roughly 0.95 to 1.28 Ga (e.g. Litherland et al., 1989; Tohver et al., 2004; Boger et al., 2005; Santos et al., 2008; Vaughan and Pankhurst, 2008; Cordani et al., 2009).
In the Sierras Pampeanas of Santiago del Estero Province, located 400 km northeast of Bonilla Complex outcrops a Cambrian magmatism is also widely represented. The Ambargasta Batholith yielded ages of 627 ± 27 Ma, 628 ± 30 Ma and 523 ± 4 Ma (Rb-Sr, Castellote, 1985, Millone et al. 2003). In hornfels of La Clemira Formation, Kouhkarsky et al.(1999) reported an age of 567 ± 16 Ma (K/Ar, whole rock). Llambías et al. (2003) determined a U-Pb age in zircon of 684+22 -14 Ma for the ignimbritic rocks of La Lidia Formation cropping out east of the Ambargasta Batholith. The rhyolitic rocks of the Cerro de los Burros yielded a Rb-Sr age of 607 ± 7 Ma (Millone et al. 2003).
Conventional U-Pb dating in zircons of the Güiraldes Thondhemite, from the Sierras Pampeanas of Córdoba Province yielded an age of 645 ± 9 Ma , whereas El Pilón Granodiorite give 707 ± 14 Ma (Rapela et al., 1998b)., Rapela et al. (2005) also dated cores of detrital igneous zircons from a para-amphibolite from the Difunta Correa metasedimentary sequence crop out in the Sierra Pampeanas de Pie de Palo. They obtained ages between 580 Ma and 620 Ma, considered as derived from Gondwanan areas. See also van Staal et al. (2011) and Mulcahy et al. (2007).
Further south, in the Sierras Pampeanas of San Luis Province, the Paso del Rey Granite was dated in 608+26 -25 Ma by von Gosen et al. (2002). The Nogolí Complex, located in western part of these sierras yielded an U-Pb age of 554 ± 4.8 Ma in gneisses (Vujovich and Ostera, 2003).
Consequently, the most proximal source of the Pampean zircons of the Bonilla Complex are the Sierras Pampeanas, located immediately east (actual coordinates). Both analyzed samples and therefore the sequences of this unit were formed by a major input of detritus from a Brazilian-Pampean-age source.
Laurentian sources of ~500-600 Ma zircons
In order to explain the Neoproterozoic igneous zircons found in the El Quemado Formation (ca. 630–550 Ma) outcropping in the Sierra de Pie de Palo, Naipauer et al. (2010) have cited the magmatic activity due to rifting of eastern Laurentia from western Gondwana during Cambrian times (Cawood and Nemchin, 2001; Cawood et al., 2001). In the area analyzed by Cawood and Nemchin (2001), the Newfoundland Appalachians, the rifting event is documented by several rift sequences including the South Brook, Summerside and Blow-Me-Down Brook formations. The first one showed only one zircon with an age of 572 ±14 Ma, whereas the second yielded ages of 586 ± 11 Ma (two zircons), 628 ± 12 Ma (two zircons) and 760 ± 40 Ma (one zircon). The Neoproterozoic detrital zircons are possibly associated with rift-related igneous activity which occurred from 760 Ma to 550 Ma (Cawood et al., 2001). As expected, the three above mentioned formations have ~ 1Ga age zircon populations appropriated for the Grenvillian basement they cover. The South Brook and Blow-Me-Down Brook formations have an important population of zircons with ages between 2.8 and 2.6 Ga. These were assigned to the large scale magmatic activity in the Superior craton and were also recognized in the Trans-Hudson, Penokean, Yavapai and Mazatzal orogens amongst others.
The known phases of silicic rift-related igneous activity within the orogeny are only from the U.S. Appalachians, located more the 2,000 km south. The Catoctin Formation outcrops from southern Pennsylvania to central Virginia (Aleinikoff et al., 1995), and consists mainly of tholeiitic flood metabasalt. At South Mountain, Pennsylvania, it is represented by 520 m of metarhyolites with interbedded metapelitic rocks (170 m). In the Blue Ridge anticlinorium, Virginia, the formation shows a lower part, 500 m thick, composed of metabasalt breccias, whereas the upper part, 2 km thick, is dominated by metabasalts, some of them referred to as subaqueous pillowed flows (Kline et al., 1987). Samples from the metarhyolites yielded ages from 564.3 ± 9.3 Ma to 801 Ma indicating inherited zircons, including ages of 648 Ma to 601 Ma. The ages from felsic dikes, that possible feed the metarhyolites yield a zircon age of 571.5 ± 4.7 Ma (Aleinikoff et al., 1995).
The Mount Roger Formation outcrops in southwestern Virginia, North Carolina and Tennessee. It is composed of 300 m thick, low-silica rhyolite lava flows of the Buzzard Rock Member, followed by a 750 m thick Whitetop Rhyolite Member that consists of high-silica, phenocryst-poor rhyolitic lava flows. The upper part, the Wilburn Rhyolite Member, is formed of 760 m of porphyritic high-silica welded ash-flow tuffs. According to Aleinikoff et al. (1995) two samples of the Whitetop Rhyolite Member yielded ages of 758 ± 12 Ma and 772 ± 24 Ma. Ages of 371, 703, 751, and 760 Ma were also obtained, but were considered as having been affected by the Carboniferous Alleghenian orogeny. All evidence points to a terrestrial origin for the Mount Rogers Formation. Therefore, the age of 760 Ma was considered by Aleinikoff et al. (1995) as the time of extrusion of the acidic magmatic event during an early episode of continental rifting (760-700 Ma) developed in the central and southern Appalachians, but without continental separation.
A younger pulse of igneous activity (570-560 Ma) led to continental separation and the opening of the Iapetus Ocean. This event is associated with the emplacement of intrusive bodies (Tollo et al. 2004), mostly of A-type, including the Yonkers Gneiss (563 ± 2 Ma), the Pound Ridge Granite (562 ± 5 Ma), and the Suck Mountain Pluton at 680 ± 4 Ma.
Consequently, the Neoproterozoic-Cambrian sequences of the central and northern Appalachians show significant extensive bimodal volcanism, up to 2,000 m thick, associated with A-type magmatism recognized along 2,600 km from Newfoundland to North Carolina. If Cuyania or Precordillera terranes come from this area we should be able to find evidence of such a vast magmatic event in the western border of Cuyania.
However, there are notable differences when comparing the stratigraphy and tectonic of the Appalachians between Newfoundland and North Carolina with coeval sequences of the Cuyania Terrane. Here, the Neoproterozoic-Cambrian time is represented by passive-margin sequences, and acidic volcanic and intrusive events have never been recognized. Moreover, tholeiitic flood basalts like those of the Appalachian area were never recognized, not in Cuyania, Precordillera, nor Cordillera Frontal. If Cuyania was attached to the Laurentia margin at that time, they must share similar magmatic events. Thus, the Laurentia margin from Newfoundland to North Carolina is not a likely source region for the protolith of the Bonilla Complex, neither its zircons.
Latest Proterozoic-Early Cambrian sedimentary rocks outcrop between Alabama and Vermont and they have been assigned to the Chilhowee Group. Between Alabama and northwestern Virginia (Walker and Driese, 1991), the lower Unicoi and Cochran formations of this group are up to 500 m thick and represent coastal braid plain deposits. Basalt flows in the Unicoi occur at ~570-555 Ma (Southworth et al., 2007). The Nichols-Hampton formations are 275 m thick and represent a silt and mud dominated marine shelf. The Nebo and Murray formations are 120 and 220 m thick respectively. The first was coeval to the Nichols-Hampton formations and was deposited in storm-dominated inner shelf, shoreface and foreshore environments.
The second unit represents a low-energy mud shelf. The Hesse (100 m thick), Helenmode (60 m thick) and Shady formations form the upper part of the Chilhowee Group. They were deposited in shoreface and foreshore environments of a clastic-dominated shelf transitional to a carbonate ramp. A Rb-Sr recalculated age of 539 ± 30 Ma was obtained in glauconitic samples of the Murray Formation, which reinforces the Vendian to Early Cambrian age estimations based on determinations of trilobites, ostracodes, acritarchs, etc.
The Chilhowee Group then represents a fluvial-to-marine, late synrift to early drift succession that unconformably covers the Grandfather Mountain Formation considered Neoproterozoic (800-900 Ma). This unit is up to 9 km thick and represents alluvian-fan and braided river deposits (Schwab, 1977).
Such association of sedimentary environments has also never been recognized in the western border of the supposed Cuyania Terrane. Therefore, no evidence supports the idea that the Precordillera or Cuyania Terranes were located adjacent to the North American craton between Alabama and Virginia at that time.
Further south, the Ouachita embayment was considered by Thomas et al. (2000) as the location of the rifting of the Precordillera Terrane. The time of the rifting event was constrained by the synrift rhyolites of the Arbuckle uplift in southern Oklahoma, with U-Pb zircon ages of 536 ± 5 Ma and 539 ± 5 Ma, also consistent with the age of the igneous rocks in the Wichita Mountains.
Gilbert (1983), McConnell and Gilbert (1990), and Puckett (2011) analyzed the igneous stratigraphy of the Southern Oklahoma Aulacogen, where the Arbuckle uplift is located. The Oklahoma Aulacogen is part of a series of rifts developed at high angles to the southern margin of the North American craton and extending from Oklahoma to Utah.
In the Southern Oklahoma Aulacogen the first pulse of magmatism, which appears on the ~1.2- 1.4 Ga basement, is represented by the ~600-577 Ma (Lambert et al., 1988) rocks of the Glen Mountains Layered Complex (anorthositic-gabbroid rocks) of the Raggedy Mountain Gabbro Group.
Basaltic volcanic rocks (320 m thick) of the(~550 Ma) seem to be associated with the intrusive event. These rocks are compositionally tholeiitic and the basalts were considered to be shallow submarine flows. Small later plutons (Roosevelt Gabbros) are intruding (552 Ma, U-Pb zircon, Bowring and Hoppe, 1982) the igneous sequence. The intrusion of this unit seems to be associated with uplift and erosion of older units, which were later covered by the Carlton Rhyolite Group (~525 Ma), associated with the intrusion of A-type acidic magmatism of the Wichita Granite Group. The rhyolites are up to 1,400 m thick and it is believed that they erupted along linear fissures (McConnell and Gilbert, 1990). McCleery and Hanson (2010) indicated more than 2 km thick and A-type magmatism characteristics typical of emplacement in rift settings.
According to Gilbert (1983), the volume of these rocks can easily reach 40,000 km3. Sedimentation continued in marine carbonate platforms, where the Cambro-Ordovician Timbered Hill and Arbuckle groups were deposited. A deep borehole (5,640 m, Puckett, 2011) cut 3,657 m of basaltic rocks and 650 m of rhyolites. The lower part of the borehole (470 m) showed interbedding of basalts, rhyolites, dolomites and lithic dolomitic conglomerates. The basaltic rocks were correlated with the Navajoe Mountain Basalt-Spilite Group.
Hanson et al. (2009) interpreted these rifting-related rocks as occurred during the opening of the Iapetus Ocean, either as part of an r-r-r triple junction or within a leaky transform at an offset in the Laurentian margin.
The orientation of the Southern Oklahoma Aulacogen suggests a transform fault that propagated into the Laurentian continental crust. The fault system is nearly parallel to the Alabama-Oklahoma Transform, which was hailed as the northern border of the supposedly migrated Precordillera terrane (Thomas, 2011).
If this terrane was adjacent to the Ouachita border of the North American craton we could expect a similar transform fault system in the northern border of Precordillera, or at least evidence of such a noteworthy Cambrian magmatic event.
Further south, in the Llano and Devils River uplift, Texas, the metasedimentary-metavolcanic succession is 850 m thick, including basaltic rocks (529 ± 31 Ma, Rb-Sr, Nicholas and Rozendal, 1975). The last is located near the Texas Transform, which was referred to as the southern border of the Precordillera Terrane.
The Thomas (2011) palinspastic reconstruction of the Iapetan rifted margin of southern Laurentia, shows the location of the Precordillera Terrane between the Alabama-Oklahoma and Texas transforms, extending more than 850 km in a north-south direction and 830 km in an east-west direction, with important basic and acidic magmatism near these transforms.
The supposed Cuyania Terrane is indeed more than 1,100 km long in a NNW-SSE direction if the Ordovician limestones of La Pampa Province are included, and 200 km wide in an east-west direction. Even if the E-W pre-Andean and Andean deformation is restored by a 55% (average using the values of Giambiagi et al., 2009), we are far away from the 800 km required by the Thomas (2011) model. Another possibility is to consider that part of the Cuyania Terrane extends below the Andes Cordillera until the Chilean coast of the actual Pacific Ocean. However, again there are considerable differences with the form and size of the model proposed by Thomas (2011). The northern and southern borders of the supposed Cuyania Terrane, the Alabama-Oklahoma and Texas transforms are not represented in the northern and southern tips of the area considered as the Cuyania Terrane, neither a trace nor relict of such structures were found. Another discrepancy of the Thomas (2011) model is the presence of the Southern Oklahoma Aulacogen, which propagates for more than 1,000 km into the North American craton. This structure and its conspicuous magmatism must be represented in the Cuyania Terrane because it was supposedly part of the North American craton during Neoproterozoic-Early Cambrian times. However, such elements were never found in Precordillera or Cuyania Terrane.
According the evidence discussed above, it is difficult to reconcile the idea that the zircons and the protolith of the Bonilla Complex were derived from the Ouachita area in Texas. Most evidence indicates a Gondwana origin for the zircons and the material that form the Bonilla Complex. It is not necessary to consider the presence of an exotic terrane between the Bonilla Complex and the Gondwana margin in order to explain the ~ 1Ga zircon recorded in the first. Such a population of zircons was also found widespread in many Pampean units of the Sierras Pampeanas, the Puncoviscana Formation and in northern Patagonia, but exotic terranes, located east or northeast of such outcrops, was not proposed to explain such a population.
Post-passive margin evolution of the Bonilla Complex
The development of the Bonilla Complex as a passive margin probably continued until the Dapingian (471 Ma, Middle Ordovician), see Figure 11c. As previously indicated, Ordovician to Devonian rocks with graptolites and flora fossil remains (Cuerda et al., 1988; Cortés, 1992) were found in the Villavicencio Formation and Cienaga del Medio Group, which outcrop immediately west, east and north of the area studied.
The age of the deformation of the Bonilla Complex and equivalent units were established by Buggisch et al. (1994), who dated illite/muscovite possibly related to this event in the metasedimentary rocks of the Bonilla area, determining an age of 437 Ma (K/Ar). The slices of the oceanic crust tectonically interbedded into the metasedimentary rocks of the Bonilla Complex rocks were emplaced and deformed coevally with their host rocks at 418 ± 10 Ma (Davis et al. 2000), see Figure 11c. The facies association documented in the Bonilla Complex can be assigned to an accretionary complex, similar to that recognized by Gregori and Bjerg (1997) in the Metales Belt of the Cordillera Frontal.
Buggisch et al. (1994) also determined an age of 353 Ma in the Bonilla Complex, whereas Davis et al. (1999), using Ar/Ar, reported ages of 384 and 385 and 372 and 377 Ma in the metamorphic rocks of the Portillo area. These ages are grouped between the Lower Mississippian and the Frasnian (Upper Devonian) and seem to represent the Chanic deformation, widely recognized in western Argentina (Fig. 11c).
The Cuyania Terrane collision
As suggested by several authors (Ramos et al., 1998; Ramos, 2004; and references therein), the Cuyania Terrane collided with the Gondwana margin in the Middle Ordovician (465 to 454 Ma) or Middle–Late Ordovician (Thomas and Astini, 2003). The first ages are represented by the assumed syncollisional and postcollisional Famatinian magmatism in the Sierras Pampeanas.
The Famatinian magmatic event (Lower Ordovician-Late Devonian), supposedly due to the approach and collision of the Cuyania Terrane with the Pampean (Fig. 11c) margin of the Gondwana continent, extends from north to south between the Sierra de Quilmes, Aconquija, Belén, Ancasti, Ambato, Velasco, Chepes, Valle Fértil, La Huerta, San Luis, La Pampa and northern Patagonia (Toselli et al., 2003; Dahlquist and Baldo, 1996; Hauzenberger et al., 2001; Büttner et al., 2005; Otamendi et al., 2008, Delpino et al., 2007, Dahlquist et al., 2008). Some authors prolong this orogeny into Bolivia and Perú by more than 3,500 km (Collo et al., 2009).
The extensive Rb-Sr, K-Ar and U-Pb dating of magmatic rocks in the Sierras Pampeanas and Sierras de Famatina show that the development of this magmatism extends between ~488 and ~450 Ma (Fig. 11c). Specifically in the last region, the ages are partially overlapped by the development of the Famatinian basins (477-463 Ma) as indicated by Dahlquist et al., (2008). The magmatic arc is coeval with the sedimentary basin, being indicative of extensional conditions between ~488 and ~450 Ma. According to Alonso et al (2008) extensional deformation prevailed during Middle to Upper Ordovician times (~466-443 Ma) in the Don Polo, Alcaparrosa and other formations of western Precordillera.
These evidences, as well as the continental extension of the Famatinian magmatic event, is in conflict with the hypothesis that the Cuyania Terrane collided against the Gondwana margin between 465 and 454 Ma, and precludes consideration of its collision being responsible of the magmatic event.
The limestones of the Bonilla Complex, Precordillera and Sierras Pampeanas
One of the most noted differences between the Cuyania Terrane and the rest of the Gondwana margin (Sierras Pampeanas, Famatina, Cordillera Oriental) is the abundance of Cambrian-Ordovician limestones in the former and their near absence in the others. Accordingly, the abundance of limestones in the Cuyania Terrane and in the southern and eastern margin of Laurentia was considered evidence of a similar origin.
As indicated above, the limestones represented by lithofacies III, up to 1 km thick, are interbedded into the metasedimentary sequence of the Bonilla Complex. These rocks can be successfully correlated with the carbonate metasilstone sequence of the Cortaderas Formation (Cucchi, 1972) and with the Alojamiento Formation, outcropping 40 km north of our studied area. The latter was assigned to the Middle-Upper Cambrian (Bordonaro, 2003). Equivalent units in the eastern Precordillera include the La Flecha, Zonda and La Laja formations.
Linares et al. (1982), Thomas et al. (2001), Buggisch et al. (2003), Galindo et al. (2004), Naipauer et al. (2005), Sial et al. (2008) and Murra et al. (2011) carried out carbon, oxygen, 87Sr/86Sr, and other trace element isotope analyses in carbonate rocks of the Precordillera and the Sierras Pampeanas.
Thomas et al. (2001), reported 87Sr/86Sr values for gypsum of the Early Cambrian Cerro Totora Formation in the Precordillera and the Rome Formation of the Birmingham Graben in the Ouachita embayment. The values are between 0.70877 and 0.70867, indicative of an Early Cambrian (~524-~510 Ma) marine restricted environment.
Galindo et al. (2004) analyzed samples from Los Hornos and San Juan formations of the Precordillera, the Caucete Group, the Filo del Grafito marbles, the Difunta Correa Sequence, and the Ophiolitic unit of the Sierra de Pie de Palo, and marbles of the Pan de Azucar, located in the Sierras Pampeanas of the San Juan Province. They also analyzed samples from the Sierra del Gigante in the San Luis Province, part of the Sierras Pampeanas. An important result of their study is that the Difunta Correa Sequence cannot be correlated with carbonate rocks of the Precordillera and the Caucete Group, but the 87Sr/86Sr and the d13C results are coincident with the values of marbles of the Sierra de Umango (Varela et al., 2001). The 87Sr/86Sr results for the Caucete Group (0.709017 to 0.711032) indicate strong similarities with the Filo del Grafito marbles and Los Hornos Formation (0.708813, Early to Middle Cambrian, 542-500 Ma) Furthermore, isotopic values of rocks of the Sierra del Gigante show remarkable similarities with those of the Caucete Group and Filo del Grafito marbles (Galindo et al., 2004). After a detailed discussion about the absence of Neoproterozoic rock equivalents to the Difunta Correa Sequence in the southern Laurentian margin, it was concluded that the Difunta Correa Sequence and its Mesoproterozoic basement can be autochthonous or para-autochtonous to Gondwana.
Naipauer et al. (2005) reported results on carbon and oxygen isotopes, 87Sr/86Sr, 206Pb/207Pb and Sm/Nd ratios from rocks of the Angacos Limestone and the limestones of Cerro Salinas in the Sierra de Pie de Palo. They also studied the carbonates of Loma de Las Chacras, and the Pan de Azucar Marble near de Sierra de Valle Fértil. From the Precordillera they analyzed the Cambrian La Laja Formation. For the Angacos Limestone the values of d13C are equivalent to those of Linares et al. (1982) and Buggisch et al. (2003), and equivalent with values of the La Laja Formation. The 87Sr/86Sr values (0.70884 and 0.70896) of the La Laja Formation are consistent with a Middle Cambrian age, whereas the limestones of Cerro Salinas are possibly upper Middle to Upper Cambrian (0.70915-0.70993).
Murra et al. (2011) analyzed the 87Sr/86Sr ratios in metacarbonates of the Sierra Brava Complex (Aceñolaza et al., 1983) in the Sierra de Ancasti (Sierras Pampeanas de Catamarca). According to Murra et al. (2011), two populations can be differentiated; one possibly deposited in the Ediacaran, (0.70745 to 0.70787) between 635 and 560 Ma and the other in the Lower Cambrian, the first being comparable with those of the Sierras Bayas and the Arroyo del Soldado, from the Río de la Plata Craton. Because they can also be correlated with the Difunta Correa Sequence of Sierra de Pie de Palo and the marbles of the Sierra de Umango, it seem that all have a common origin.
Although the Lower Cambrian population (0.70831 to 0.70867) is smaller, and according to Murra et al. (2011) they probably underwent post-sedimentation alteration, they cannot be disregarded because the extensive outcrops of carbonates interbedded in Ediacaran to Cambrian sequences (Puncoviscana and equivalent formations) of the Sierras Pampeanas. Indeed, these results are curiously similar to those obtained for the Las Tienditas Formation, which is interbedded into the Puncoviscana Formation, as well as being similar to the Los Hornos Formation and the Caucete Group of the supposed Cuyania Terrane.
Misi et al. (2007) presented isotopic results for the Neoproterozoic-Cambrian carbonate rocks of the South American platform, including the Bambuí Group (740± 22Ma, Babinski and Kaufman, 2003), the Vazante Group (Latter Neoproterozoic, Azmy et al., 2001), the Una Group (600-750 Ma, Misi and Veizer, 1998), the Corumbá Group (550- 504 Ma, Misi et al., 2007), the Arroyo Soldado Group (633-532 Ma, Hartmann et al., 2002; Kawashita et al., 1999), as well as the Sierras Bayas Group of Tandil in the Rio de la Plata Craton (Gómez Peral et al., 2007). From these, the 87Sr/86Sr results for the Corumbá Group (0.70852, Boggiani et al., 2003) and the Serra do Garrote Formation Vazante Group (0.70869-0.70886) are equivalent to those of the Los Hornos, Cerro Totora and La Laja formation of the Precordillera, indicative of a Gondwana affinity.
The limestones and marbles intercalate into the Puncoviscana or equivalent formations can be recognized all along the Sierras Pampeanas. In the Loma Corral Formation (Sierra de Fiambalá, 27° 20´ S-67° 15´ W, Catamarca Province) there are outcrops of limestones and marbles, up to 1000 m thick, interbedded into low-degree metamorphic rocks that were correlated with the Puncoviscana Formation (Ruiz Huidobro, 1975). In the Sierra de Altohuasi, located 50 km northeastern (26° 57´ S-66° 53´ W) of the above mentioned outcrops, the Totoralilla Formation, constituted of marbles 400 m thick, 1 km long is interbedded into the Loma Corral Formation (García, 1974).
In the Sierra del Aconquija (Sierras Pampeanas, Tucuman Province, 26° 36´ S-65° 40´ W) there are outcrops of the Peñas Azules Limestones (up 300 m thick) interbedded with the Puncoviscana Formation (Gonzalez et al., 2000). Further north, in the eastern border of Puna, there are outcrops of limestones of the Cambrian Pachamama Formation (Hong et al., 2001).
Near Pomán, in the Sierra de Ambato (28° 26´ S-66° 13´ W, Catamarca Province) interbedded in schists and phyllites correlated with the Puncoviscana Formation, appear limestones up to 10 m thick. Other outcrops of limestones and marbles interbedded in equivalent units of the Puncoviscana Formation in the Catamarca Province appear in Muschaca (27° 30´ S-66° 20´ W), and Esquiú (29° 18´ S-65° 22´ W).
In the La Rioja Province, the outcrops are located in the Sierra de Umango (29° 05´ S-68° 42´ W) near Villa Unión (29° 17´ S-68° 17´ W) and in the Sierra Brava (29° 54´ S-65° 48´ W). The distribution of the outcrops of limestones and marbles interbedded in Neoproterozoic-Cambrian low-degree or metasedimentary rocks in the Sierras Pampeanas of Catamarca and La Rioja alone indicate an area of depositation more than 400 km long in a north-south direction and 350 km wide in a west-east direction.
In the Sierras Pampeanas of Córdoba and San Luis Provinces, the outcrops of limestones and marbles interbedded in Neoproterozoic-Cambrian low-degree or metasedimentary extend more than 325 km in a north-south direction and 170 km in a west-east direction. The main outcrops are located at Quilpo (30° 49´ S-64° 41´ W), Cunuputu (30° 52´ S-64° 40´ W), Ischilín (30° 39´ S-64° 20´ W), Sierra Chica (30° 56´ S-64° 24´ W), Candonga (31° 04´ S-64° 19´ W), Yulto (33° 15´ S-65° 32´ W), Achiras (33° 08´ S-64° 59´ W), and Alta Gracia (31° 46´ S-64° 27´ W).
There are also outcrops in the La Pampa Province (400 km south of the above mentioned site) including the Cerro Rogaziano marbles, the San Jorge Formation limestones (37° 27´ S-66° 21´ W), and scattered outcrops near Sierra de Lihuel Calel. However, these rocks were assigned to the Lower Paleozoic (Melchor et al., 2000) placing doubt on the correlation with the rocks of the Sierras Pampeanas.
Further south, in northern Patagonia, marbles up to 1 km thick are interbedded in the Mina Gonzalito Complex (535–540 Ma, Pankhurst et al., 2006) and in the El Jagüelito Formation (∼535 Ma, Pankhurst et al., 2006). Indeed, in the El Jagüelito Formation the trace fossil content is comparable with those of the Puncoviscana Formation (González et al., 2002), suggestive of an Ediacaran-Cambrian depositional age and for the continuity of the marine basin where the Puncoviscana and equivalent units were deposited from the Sierras Pampeanas to northern Patagonia.
Therefore, as indicated above, Neoproterozoic to Cambrian limestones and marbles are common all along of the 2,000 km of the Pampean margin of the Gondwana-South America continent during the Neoproterozoic-Cambrian time, not only in the supposed Cuyania Terrane.
The provenance analyses of the Bonilla Complex indicate that its protolith was derived from an older exhumed felsic basement belonging to an upper continental crust.
No evidence of basic or acidic volcanism of continental rift type was found in the Bonilla Complex or equivalent units in western Precordillera, nor in the areas where the northern and southern borders of the Cuyania Terrane must be located. This is in conflict with the presence of thick sequences of such rocks in the Ouachita area of the Laurentia margin.
The detrital zircons from the Bonilla Complex place the maximum age of sedimentation at ca. 592 Ma. The distribution of the zircon populations are equivalent to those of the Puncoviscana, Ancasti, La Cébila, and Tuclame formations of the Cordillera Oriental and the Sierras Pampeanas. Equivalent units in the Sierras de Córdoba and San Luis yield similar results. They are also comparable with those of the El Jagüelito and Nahuel Niyeu formations of northern Patagonia. It is not necessary to suggest the presence of an allochthonous terrane between the Bonilla Complex and the Gondwana margin to explain the 1 Ga zircon population.
The limestones of the Bonilla Complex, deposited in a passive margin, can be correlated with Cambrian limestones of Precordillera, Sierras Pampeanas and northern Patagonia.
Therefore, the source of the Bonilla Complex protolith was located east (actual coordinates) and the rocks were deposited in a passive margin along the western coast of Gondwana during Neoproterozoic-Cambrian time, covering a basement that was already part of the Gondwana continent at that time.
Por otro lado, evidencias convincentes de que las rocas del basamento no son de Laurentia provienen de 1) edades neoproterozoicas tardías a cámbricas inferiores en circones de las areniscas cámbrica media del olistolito San Isidro
en el Estancia San Isidro (Precordillera mendocina), y 2) Mesoproterozoico inferior
1600-1500 Ma en las areniscas cámbricas medio
del Miembro Soldano de la Fm La Laja, en las areniscas del Ordovícico medio del tope de la Fm Estancia San Isidro y en cantos rodados ígneos del Ordovicico Superior de la Fm
Empozada y 3) circones del Neoproterozoico en
muchas areniscas cámbricas-ordovícicas (Finney et
al., 2005b, 2005c; Gleason et al., 2007).
Arenisca del Olistolito San Isidro
Es una muestra de arenisca de la base del olistolito San Isidro, uno de los grandes en la parte más baja del Ordovícico medio de la Fm Estancia San Isidro de Heredia y Beresi (2004) y que Keller (1999) y Astini (2003) consideraron representante de la Fm. Los Sombreros. Este olistolito se puede seguir a lo largo de mas de 5.6 km y su estratigrafía se ha descripto en varias secciones (ej., Quebrada, de San Isidro, Quebrada Agua del la Cruz) (Bordonaro et al., 1993; Keller et al., 1993; Keller, 1999,; Heredia y Beresi, 2004) que muestran que las arenisca está en continuidad estratigráfica con caliza del Cámbrico medio con trilobites. De este modo se considera que las areniscas se habrían depositado en la plataforma de calizas Cuyania durante el periodo de tiempo que siguió a la separación de Cuyania de Laurentia y durante su historia temprana de deriva y hundimiento térmico en el margen pasivo según el modelo Laurentiano
De acuerdo a la edad de depositación de las areniscas (~512-500 Ma), , la edad de sus circones (615-511 Ma), la textura (pobremente ordenada, granos gruesos y angulares) , y su mineralogía (arcosas con escaso Fk) requiere 1) que la arenisca fue erosionada directamente del basamento de Cuyania, 2) que las rocas del protolito fueran plutónicas tonalíticas a trondhjemiticas, que típicamente forman arcos magmáticos, 3) que estos plutones hayn intruido a lo largo de 100 Ma, y 4) que inmediatamente después de la intrusión, estos plutones hayan sido levantadas y erosionadas para producir el sedimento que era depositado en el mar poco profundo que cubre la plataforma calcarea de Cuyania.
¿ Se puede reconciliar la existencia de un arco magmático de 100 Ma de duración durante el Neoproterozoico tardío-Cámbrico medio y su levantamiento inmediatamente subsecuente y erosión en el Cámbrico medio con el episodio de rifting y subsidencia posterior del modelo Laurentiano?
Es improbable, pero éso es lo que se requiere si se incorpora la evidencia de las areniscas del Olistolito San Isidro en el modelo Laurentiano.
¿Pueden las composiciones tonalíticas y trondhjemiticas y de 100 Ma de duración estar asociados o ser comparables con las composiciones gabroides/basálticas y graníticas/riolíticas de 10 Ma de duración en la provincia ígnea Wichita (Thomas y Astini, 2003), supuestamente ubicada en el margen del embayment de Ouachita de donde Cuyania se separó?
¿Puede la virtual ausencia de circones Grenvillianos en la arenisca San Isidro ser explicada si Cuyania se separó desde la bahía de Ouachita, teniendo en cuenta la asunción que las rocas del basamento de Cuyania son de edad Grenvilliana y de afinidad Laurentiana y dado que los cristales de circón detrítico de las areniscas Ordovícicas y Silúricas depositadas en el embayment de Ouachita, sur de Laurentia son de edad Grenvilliana (Gleason et al., 2002)?
Muestras del Miembro Soldano, Fm La Laja y otras muestras relacionadas
Intercalados en las calizas del Miembro Soldano se depositaron areniscas en la plataforma durante el Cámbrico medio. La mayoría de los circones tienen edades 1600-1500 Ma (Fig. 4), justo coincidente con un tiempo de inmovilidad magmática y tectónica, a lo largo de todo Laurentia salvo una área pequeña en Labrador (Karlstrom et al., 2001; Ross y Villeneuve, 2003). A pesar de la madurez textural y mineralógica de estas areniscas, los circones tienen una población unimodal con una edad muy antigua respecto al sediemnto que los contienen indicando que se trata de areniscas de primero ciclo, erosionadas de basamento adyacente a Cuyania durante Cámbrico medio.
Un caso idéntico, de población unimodal de 1600-1500 Ma
se encuentran en circones de areniscas del tope del Ordovícico medio de la Fm Estancia San Isidro en la Quebrada de
San Isidro (Fig. 3; Finney et al., 2005c; Gleason et al., 2007).
Más arriba en la misma sección cantos rodados (30-100 cms) de granitoides redondeados ocurren junto
con calizas y cantos rodados en arenisca, en el miembro más bajo de la Fm Empozada. Estos cantos rodados indican Mesoproterozoico temprano con
edades de 1500-1300 Ma, y un poco de granos mas viejos, heredados, con
centros de 1550 Ma (Finney et al., 2005c).
Estos tres ejemplos son evidencia definitiva que el basamento de Precordillera incluye Mesoproterozoico temprano de tipo plutónico incluye un componente sustancial que tiene una
edad de la cristalización de 1600-1500 Ma y que estas rocas fueron expuestas durante el Cámbrico medio,
Ordovícico medio y tradío.
Para reconciliar esta evidencia con la del microcontinente Laurentiano se debe concluir que
1) la región del embayment de Ouachita debió incluir durante el Cámbrico temprano roca ígneas del North
American magmatic gap, aunque tales rocas son
desconocidas allí y no se representa en circones detríticos del Ordovícico y Silúrico depositados en el embayment de Ouachita (Gleason et al.,
2002)
2) estas rocas del basamento deben permanecer soterradas y no han dejado otro rastro más que en las roca del Cámbrico y Ordovícico de la Precordillera.
3) los sistemas deposicionales Cámbrico medio se ubicaron contra las fuentes de edad Grenvilliana.
Aunque no abundante, hay circones de 700-600 Ma en las areniscas del Cámbrico y Ordovícico. Éstos
incluyen:
1) las areniscas del Ordovícico de la parte norte (Fm Las Vacas), centro (Fm La Cantera), y
del sur (Fm Empozada) de la Precordillera (Gleason
al del et., 2007)
2) clastos en el Ordovícico Superior en conglomerados (Fm La Cantera Fm), y
3) una arenisca de la parte mas baja del Miembro
El Estero de la Fm La Laja, quizás, reciclados de areniscas más viejas. De hecho,
ellas se pueden haber retrabajado de una arenisca Neoproterozoica
comparable a una fuente de los circones detríticos de la Difunta Correa, una sucesión metasedimentaria
de la Sierra de Pie de Palo (Rapela et al. 2005a).
Los circones de 700-600 Ma están
extensamente distribuidos en el Cámbrico-Ordovicico sedimentario
de Precordillera. Interesantemente, circones detríticos de
esta edad está casi ausente en el Ordovícico y Silurico depositado en el embayment de Ouachita en
Laurentia (ej., Gleason et al., 2002).
Circones detríticos de unidades de Precordillera y Sierras Pampeanas
Afinidad del basamento de Cuyania: Relaciones isotópica de Pb
Los ampliamente sostenida interpretación que
el basamento de Cuyania es de afinidad de Laurentiana está basado
no sólo en la supuesta edad Grenvilliana del basamento sinó también en las relaciones isotópicas de Pb de las roca.
Kay et al. (1996) mostró que los xenolitos en rocas volcánicas miocenas de Precordillera, y que se interpretaron que representaban muestras del basamento de edad Grenvilliana continen el Pb menos radiogenico desde el Precambrico hasta el Reciente en
América del Sur y que la signatura de estas rocas es
similar al de la Provincia Grenville de norteamerica,
sobre todo en el Levantamiento El Llano, Texas, cerca del
Embayment de Ouachita.
Fm. Alcaparrosa
Esta clásica unidad que aflora en el sector noroeste de la
sierra del Tontal, se caracteriza por un conjunto de areniscas
finas, limolitas y pelitas laminadas, silicificadas, con tonos
oscuros y claros por alteración. Aloja cuerpos tabulares,
mantos y filones de rocas básicas (ofiolitas).
El contenido paleontológico de graptolites indican una
edad llandeiliana tardía a caradociana temprana según Schauer
et al. (1987). No se conoce la relación estratigráfica con
unidades similares colindantes por sus contactos Tectónicos,
pero se interpreta que serían variaciones faciales de las
Formaciones Cabeceras y Don Polo.
La asociación de facies indican un ambiente de llanura
abisal.
Formación Villavicencio
Esta unidad, definida por Harrington (1941), comprende los depósitos de la «facies normal» del Grupo Villavicencio sensu Harrington (1971). Está integrada por capas de wackes y pelitas, de coloración verde grisácea, las que en general presentan arreglo turbidítico (González Bonorino, 1975a, b), y cuya localidad tipo se ubica en los alrededores de Uspallata.
Borrello (1969) interpreta los depósitos de esta formación como facies típica de flysch (ortoflysch) o «flysch Villavicencio», diferenciándola de la facies de pre-flysch o «vacuidad Cortadera» y «Los Alojamientos». Es posible que los afloramientos ordovícicos, referidos por algunos autores a la parte inferior de la Formación Villavicencio, sean equivalentes laterales de las Formaciones Cortaderas, Cabeceras, Portezuelo del Tontal y Alcaparrosa. Estas unidades presentan como rasgo común, intrusiones básicas y ultrabásicas, en partes metamorfizadas, interpretadas como cortejos ofiolíticos (Borrello, 1969; Haller y Ramos, 1984).
La Formación Villavicencio, en su localidad tipo, consiste en bancos de hasta 1 m de espesor de wackes y pelitas, de coloración verdosa a verde grisácea, con típica estructura turbidítica. En general las areniscas presentan gradación normal, base neta o erosiva con marcas subestratales de corriente y herramientas, e hypichnias, y techo gradacional a las pelitas.
Cuerda et al. (1988b), reconocen estos depósitos con la denominación de Formación Canota, la que interpretan sobrepuesta en discordancia regional a depósitos ordovícicos, que denominan Formación Villavicencio. Kury (1993) señala que el contacto entre las citadas unidades es de carácter tectónico y no sedimentario, razón por la cual sigue la propuesta de Harrington (1941), equiparando los depósitos devónicos con la Formación Villavicencio, y a los ordovícicos con la Formación Empozada. En este aporte se sigue el criterio de Harrington (1971) y el ordenamiento propuesto por Kury (1993), ya que clásicamente en la literatura geológica, se ha reconocido con la denominación de Formación Villavicencio a las sedimentitas (wackes y pelitas), con restos de plantas, referidas por los citados autores sensu lato, como de edad devónica.
Estos depósitos contienen restos vegetales y trazas fósiles de la icnofacies Nereites, entre las que destacan Neonereites biserialis Seilacher, Neonereites isp., Palaeoheominthoida isp., Planolites isp., Helminthopsis isp. Tanto la base como el techo de la formación se desconocen, debido a que está limitada por fallas (Harrington, 1971; Kury, 1993), que la relacionan con unidades del Ordovícico, estimándose su espesor aflorante en unos 2.000m (Cuerda et al., 1987).
El contenido paleontológico de la formación está representado por restos de plantas, trazas fósiles y palinomorfos, destacándose la falta de registro de shelly facies. En este contexto se destaca el hallazgo de quitinozoos por Pöthe de Baldis e Ichazo (1987), en la quebrada del río Santa Clara, extremo sur de la sierra del Tontal, compuesta por Conochitina gordonensis y Linochitina cingulata serrata, que indican una edad llandoveriana a, probablemente, ordovícica superior.
En la quebrada de San Isidro, Cuerda et al. (1987) registran restos de plantas vasculares, que atribuyen al género Barawanathia, para los que interpretan una edad devónica inferior. Sin embargo, un estudio de revisión posterior de los aludidos restos de plantas (Cuerda et al., 1993), permite concluir que los mismos indicarían una edad silúrica superior (Pridoliano). En estos depósitos, Cingolani (1970, en Cuerda et al., 1993) menciona trazas fósiles de la icnofacies de Nereites, y restos de euryptéridos.
Astini, R.A., 1998a. Stratigraphic evidence supporting the rifting, drifting and collision of the Laurentian Precordillera terrane of western Argentina. In: Pankhurst, R.J., Rapela, C.W. (eds.). The Proto- Andean Margin of Gondwana. Geological Society of London, Special Publication, 142, 11-33.
Astini, R.A., Benedetto, J.L., Vaccari, N.E., 1995. The early Paleozoic evolution of the Argentine Precordillera as a Laurentian rifted, drifted, and collided terrane: A geodynamic model. Geological Society of America Bulletin, 107, 253-273.
Bordonaro, O., 2003. Review of the Cambrian Stratigraphy of the Argentine Precordillera. Geologica Acta, 1(1), 11-21.
Casquet, C., Pankhurst, R.J., Rapela, C.W., Fanning, C.M.,
Galindo, C., Baldo, E., Gonzalez-Casado, J.M., Dahlquist,
J.M., Saavedra, J., 2005. The Maz suspect terrane: a new
Proterozoic domain in the Western Sierras Pampeanas. In:
Pankhurst, R.J., Veiga, G.D. (eds.). Gondwana 12: Geological
and Biological Heritage of Gondwana, Academia
Nacional de Ciencias, Córdoba, Argentina, Abstracts, p. 92.
Finney, S., Peralta, S., Heredia, S., 2005a. Where is the foreland basin at the Cuyania-Gondwana collisional zone? In: Pankhurst, R.J., Veiga, G.D. (eds.). Gondwana 12: Geological and Biological Heritage of Gondwana, Academia Nacional de Ciencias, Córdoba, Argentina, Abstracts, p. 156.
Finney, S., Peralta, S., Gehrels, G., Marsaglia, K., 2005b. The Early Paleozoic history of the Cuyania (greater Precordillera) terrane of western Argentina: evidence from geochronology of detrital zircons from Middle Cambrian sandstones. Geologica Acta, 3(4), 339-354.
Finney, S.C., Gleason, J.D., Gehrels, G.G., Peralta, S.H., Aceñolaza, G., 2003. Early Gondwanan Connection for the Argentine Precordillera Terrane. Earth and Planetary Sciences Letters, 205, 349-359.
Galindo, C., Casquet, C., Rapela, C., Pankhurst, R.J., Baldo, E.,
Saavedra, J., 2004. Sr, C, O isotope geochemistry and
stratigraphy of Precambrian and lower Paleozoic carbonate
sequences from the Western Sierras Pampeanas of Argentina:
tectonic implications. Precambrian Research, 131, 1041-
1056.
Gleason, J.D., Finney, S.C., Peralta, S.H., Gehrels, G.E., Marsaglia, K.M., 2007. Zircon and whole rock Nd-Pb isotopic provenance of Middle and Upper Ordovician siliciclastic rocks, Argentine Precordillera. Sedimentology, 54(1), 107-136.
Karlstrom, K.E., Åhäll, K-I., Harlan, S.S., Williams, M.L., McLelland, J., Geissman J.W., 2001. Long lived (1.8 – 1.0 Ga) convergent orogen in southern Laurentia, its extensions to Australia and Baltica, and implications for refining Rodinia. Precambrian Research, 111, 5-30.
Kay, S.M., Orrell, S., Abbruzzi, J.M., 1996. Zircon and whole rock Nd-Pb isotopic evidence for a Grenville age and Laurentian origin for the basement of the Precordillera terrane in Argentina. The Journal of Geology, 104, 637-648.
McClelland, W.C., Ellis, J.R., Roeske, S.M., Mulcahy, S.R.,
Vujovich, G.I., Naipauer, M., 2005. U-Pb SHRIMP igneous
zircon ages and LA-ICPMS detrital zircon ages from metamorphic
rocks between the Precordillera terrane and the
Gondwana margin, Sierra de la Huerta to Pie de Palo, northwest
Argentina. In: Pankhurst, R.J., Veiga, G.D. (eds.).
Gondwana 12: Geological and Biological Heritage of Gondwana,
Academia Nacional de Ciencias, Córdoba, Argentina,
Abstracts, p. 250.
Ramos, V.A., Dallmeyer, R.D., Vujovich, G., 1998. Time constraints on the Early Palaeozoic docking of the Precordillera, central Argentina. In: Pankhurst, R.J., Rapela, C.W. (eds.). The Proto Andean Margin of Gondwana. Geological Society, London, Special Publications No. 142, 143-158.
Ramos, V.A., Keppie, J.D., 1999. Laurentia-Gondwana Connections before Pangaea: Geological Society of America, Special Paper, 336, 276 pp.
Rapela, C.W., Pankhurst, R.J., Casquet, C., Baldo, E., Saavedra, J., Galindo, C., Fanning, C.M., 1998. The Pampean Orogeny of the southern proto-Andes: Cambrian continental collision in the Sierras de Córdoba. In: Pankhurst, R.J., Rapela, C.W. (eds.). The Proto-Andean Margin of Gondwana. Geological Society of London, Special Publication, 142, 181-218.
Rapela, C.W., Pankhurst, R.J., Casquet, C., Fanning, C.M., Galindo, C., Baldo, E., 2005a. Datacion U-Pb SHRIMP de circones detriticos en paranfibolitas neoproterozoicas de las secuencia Difunta Correa (Sierra Pampeanas Occidentales, Argentina), Geogaceta, 38, 227-230.
Rapela, C.W., Fanning, C.M., Baldo, E.G., Dahlquist, J.A.,
Pankhurst, R.J., Mura, J.A., 2005b. Coeval S- and I-type
granites in the Sierra de Ancasti, Eastern Sierras Pampeanas,
Argentina. In: Pankhurst, R.J., Veiga, G.D. (eds.). Gondwana
12: Geological and Biological Heritage of Gondwana,
Academia Nacional de Ciencias, Córdoba, Argentina,
Abstracts, p. 307.
Thomas, W.A., 1991. The Appalachian-Ouachita rifted margin
of southeastern North America. Geological Society of
America Bulletin, 103, 415-431.
Thomas, W.A., Astini, R.A., 1996. The Argentine Precordillera:
A traveler from the Ouachita embayment of North American
Laurentia. Science, 273, 752-757.
Thomas,W.A., Astini, R.A., 1999. Simple-shear conjugate margins of the Argentine Precordillera and the Ouachita embayment of Laurentia. Geological Society of America Bulletin, 111, 1069-1079.
Thomas, W.A., Astini, R.A., 2003. Ordovician accretion of the Argentine Precordillera terrane to Gondwana: a review. Journal of South American Earth Sciences, 16, 67-79.
Thomas, W.A., Astini, R.A., 2005. Vestiges of an Ordovician
Ocloyic west-vergent, thin-skinned thrust belt in the Argentine
Precordillera, southern central Andes. In: Pankhurst,
R.J., Veiga, G.D. (eds.). Gondwana 12: Geological and Biological
Heritage of Gondwana, Academia Nacional de Ciencias,
Córdoba, Argentina, Abstracts, p. 350.
Thomas, W.A., Astini, R.A., Mueller, P.A., Gehrels, G.E., Wooden, J.L., 2004. Transfer of the Argentine Precordillera terrane from Laurentia: Constraints from detrital-zircon geochronology. Geology, 32(11), 965-968.