NEW U-PB GEOCHRONOLOGICAL INSIGHTS INTO THE PROTEROZOIC
TECTONIC EVOLUTION OF NORTHWESTERN SOUTH AMERICA: THE MESONEOPROTEROZOIC PUTUMAYO OROGEN OF AMAZONIA AND
IMPLICATIONS FOR RODINIA RECONSTRUCTIONS
by
Mauricio Ibanez-Mejia
A Prepublication Manuscript Submitted to the Faculty of the
DEPARTMENT OF GEOSCIENCES
In Partial Fulfillment of the Requirements
for the Degree of
MASTER OF SCIENCE
In the Graduate College
THE UNIVERSITY OF ARIZONA
2010
New U-Pb geochronological insights into the Proterozoic tectonic evolution of
northwestern South America: The Meso- to Neoproterozoic Putumayo Orogen of
Amazonia and its implications for Rodinia reconstructions
Mauricio Ibanez-Mejia
Abstract
Outcrops of Meso- Neoproterozoic crust in northwestern South America are restricted to
isolated exposures of basement inliers included within the northern Andes of Colombia.
However, evidence for the existence of an autochthonous Grenville-age belt in northern
Amazonia that has not suffered significant disturbance during Andean orogenesis has not
been recognized so far. The present contribution reports ~1200 new single-zircon U-Pb
geochronological results from Proterozoic rocks of northwestern South America collected
from Andean cordilleran inliers, drilling-core samples from the foreland basin basement,
and outcrops of cratonic Amazonia in eastern Colombia (western Guyana shield). We
document the existence of a previously unrecognized Meso- to Neoproterozoic Grenvilleage orogenic belt buried under the north Andean foreland basins, herein termed the
Putumayo Orogen, and discuss its implications for Proterozoic tectonic reconstructions
involving the northwestern South American margin. Based on the interpretations of our
data, we propose a two-stage tectonometamorphic evolution for this orogenic segment
characterized by amphibolite-facies metamorphism and melting between ca. 1.05 and
1.01 Ga by inferred arc-continent collision during amalgamation of a peri-Amazonian
fringing-arc system to the continental margin, shortly followed by final continentcontinent collision and granulite–facies metamorphism at ~0.99 Ga. Geochronological
and paleogeographic constraints suggest collisional interactions with the Sveconorwegian
province of Baltica as a feasible tectonic scenario. Previous to the early collisional stage
and inception of the Putumayo Orogen, a fringing-arc system was developed outboard of
Amazonia’s leading margin during Mesoproterozoic times from ~1.3 to 1.1 Ga. It was
within this hypothetical parautochthonous arc system were the protoliths for metaigneous
and metavulcanosedimentary units found in the inliers of Colombia and Mexico could
possibly have originated, defining a commonly evolving Oaxaquian-Colombian fringing-
1
arc system prior to Rodinia’s final assembly. This interpretation finds further support in
the analysis of ~500 detrital zircons from Grenville-age metasedimentary units of the
Colombian Andes which reveal that protolith sedimentation occurred during late
Mesoproterozoic times and was exclusively derived from erosion of Mesoproterozoic
sources. Contrasting differences with detrital zircon age spectra of metasediments
recovered in core samples from the foreland basement, specifically the paucity of
Paleoproterozoic detritus sourced from cratonic Amazonia, suggests a sedimentary
disconnect between Mesoproterozoic arc-related basins and those developed on the
continental margin shortly before or relatively synchronous with collisional interactions
affecting the Putumayo orogen.
Keywords:
South
America,
Amazonia,
Putumayo
Orogen,
Rodinia,
U-Pb
Geochronology
1. Introduction
The recognition of mobile belts and tectonometamorphic provinces in Precambrian
cratons is key for the study and accurate geological reconstruction of ancient orogenic
belts, where sea-floor magnetic stripes are no longer available and reliable paleomagnetic
data for specific blocks and geological time periods is limited. For these reasons,
reconstructions of Proterozoic supercontinents and paleogeographic connections between
different cratons heavily rely on geological correlations drawn upon finely resolved
geochronological data and isotope geochemistry signatures. However, any attempt made
to reconstruct such old orogens has to deal with the problems of younger crustal
reworking, extensive sedimentary cover, and marginal terrane processes that rift
fragments of continental crust to redistribute them amongst other cratons as allochtonous
blocks.
The Amazon Craton is the largest of the Precambrian blocks that constitute the South
American platform (Tassinari and Macambira, 1999), and its role in the supercontinent
cycle during the history of the Earth has for long been inferred (Cordani et al., 2009).
However, there are still sizable areas that remain poorly or totally unknown as a result of
2
dense vegetation cover, heavy tropical weathering, and the development of Phanerozoic
intracratonic or Andean-related sedimentary basins (Figure 1). Many reconstructions for
the early Neoproterozoic supercontinent Rodinia predict collisional interactions between
a Laurentia-Baltica-Amazonia triple joint along Grenville-age (sensu lato) orogenic belts
(e.g. Dalziel, 1991; Hoffman, 1991 or Li et al., 2008 for a recent review) but precise
correlations between the dismembered orogenic segments are still a matter of debate and
are specially undermined by the particular lack of geochronological and petrologic
information from key regions of Amazonia like its northwestern margin. The north
Andean foreland basins, comprising part of northern Venezuela, eastern Colombia,
eastern Ecuador and eastern Peru (Figure 1), represent an area in excess of 800.000 km2
of covered cratonic crust that is still very poorly know, and with the exception of some
well data presented by Kovach et al. (1976) and Feo-Codoecido et al. (1984) nothing else
is know about the geochronology of this vast area.
The main objective of this paper is to present new U-Pb geochronological results from
Proterozoic basement rocks of northwestern South America, along a NW-SE transect that
covers most of the width of southern Colombia, ranging from exposures of cratonic
Amazonia to reworked blocks included in the Andean orogen (Figure 2). In addition to
new data acquired from the Colombian cordilleran inliers and the western Guyana shield
area, we present the first zircon U-Pb ages obtained from basement drilling-core samples
of the north Andean foreland basins. Our results comprise the first geochronological
evidence for the existence of a previously unrecognized segment of late Meso- to early
Neoproterozoic (Grenville-age) orogenic belt underlying these basins, herein termed the
Putumayo Orogen, and propose a distinct geological evolution with respect to its direct
Amazonian analog the Sunsas-Aguapei orogen. Recognition of this belt provides a
cratonic reference frame that places a link between the Andean cordilleran inliers and
autochthonous Amazonia basement, and permits further comparisons with other
supposedly peri-Gondwanan blocks with proposed Amazonian ancestries or affinities.
The word “Putumayo” makes reference to the name of a foreland basin in the Northern
Andes of Colombia (Figure 2), from which the well samples presented in this study were
3
retrieved. This basin is termed after the name of the river that delimits most of the
Colombian-Peruvian political border, and runs east from the southern Andes of Colombia
to merge with the Amazon River in Brazilian territory. The word “Putumayo” itself
comes from the Quechua, the most widely spoken language among the Andean
indigenous people of South America, and it means “gushing-” or “rapidly-flowing river”.
2. Geological Setting:
The Amazon Craton is subdivided in six different Precambrian provinces (Figure 1),
spanning over 2 Ga of Earth’s history. They are 1) the Archean Central Amazonian
Province (CA) formed during the time interval between ~3.0 - 2.5 Ga, 2) the early
Paleoproterozoic Maroni-Itacaiunas Province (MI) formed in the period from ~2.2 - 1.95
Ga, 3) the Paleoproterozoic Ventuari-Tapajos Province (VT) which evolved from ~1.95
to 1.8 Ga, 4) the Paleo- to Mesoproterozoic Rio Negro-Juruena Province (RNJ) formed
during the interval ~1.8 – 1.55 Ga, 5) the Mesoproterozoic Rondonia-San Ignacio
Province (RSI) from ~1.55 – 1.3Ga and 6) the Late Mesoproterozoic Sunsas-Aguapei
Province (SA) that evolved in the interval ~1.3 – 1.0 Ga. There are many recent reviews
dealing with the long-term evolution of the Amazon Craton and the distribution of the
different mobile belts, for a more detailed discussion the reader is referred to the works of
Teixeira et al. (1989), Tassinari and Macambira (1999), Santos et al. (2000), Santos et al.
(2008) and Cordani and Teixeira (2007). In general, it is believed that after the
amalgamation of two Archean microcontinents namely Carajás-Iricoumé and Imataca
during the Transamazonian orogeny (MI province), the remaining of the Paleoproterozoic
belts, the VT and RNJ, evolved trough the continuous accretion of juvenile intraoceanic
arcs to this proto-Amazonian nucleus. These two provinces are dominated by gneissic
terranes of granitic composition and crosscutting intermediate to felsic plutons, which
display predominantly mantle-derived isotopic signatures Tassinari and Macambira
(1999). The last two provinces located towards the southwestern margin of the craton, the
Rondonia-San Ignacio and the Sunsas-Aguapei, are the result of accretionary and
collisional tectonics characterized by arc-continent and continent-continent collisions
throughout the Mesoproterozoic, resulting in restricted episodes of crustal growth and
extensive reworking (Litherland and Bloomfield, 1981; Litherland et al. 1989; Sadowski
4
and Bettencourt, 1996; Cordani and Teixeira, 2007; Bettencourt et al., 2010; Teixeira et
al. 2010). Collisional tectonics affecting Amazonia during the late Mesoproterozoic have
been used as evidence for the participation of this craton during the amalgamation of
Rodinia.
3. The Proterozoic of NW South America:
3.1. Cratonic Amazonia:
The last outcrops of the Amazon Craton in northwestern South America can be found in
the Orinoco and Amazonas regions of eastern Colombia, western Venezuela and
northwestern Brazil (Figures 1; 2). Most of the geological mapping on the Colombian
side along with the only available geochronological study were conducted during the
PRORADAM project (acronym for “Proyecto Radargrametrico del Amazonas”) from
1974 to 1979 (see Galvis et al., 1979; Priem et al., 1982; PRORADAM, 1979). However,
no recent geochronological work has been carried out in the area and this is probably still
one of the most poorly known portions of the Precambrian basement of South America.
The craton in this region consists of deeply eroded, gneissic/migmatitic terranes of
dominantly granitic composition, interpreted as the northwestern extension of the wider
Rio Negro-Juruena and Ventuari-Tapajos provinces (Tassinari et al., 1996). In
Colombian territory, all the basement exposures on the eastern portion of the country are
collectively mapped as the “Mitu migmatitic complex” (Galvis et al. 1979; Priem et al.
1982; Gomez et al. 2007), where the dominantly felsic gneisses and migmatites yield RbSr isochron ages of ca. 1.55Ga. This basement complex is intruded by undeformed
granitoids, some of them displaying rapakivi textures like the Parguaza batholith of
western Venezuela (Gaudette et al., 1978), and is locally covered under metasedimentary
supracrustal sequences of the La Pedrera formation. Undeformed biotite granites from the
Guainia and Vaupes departments yield crystallization ages between 1.55 and 1.48Ga,
obtained by multigrain TIMS U-Pb analyses in zircon (Priem et al., 1982). In some areas,
predominantly
in
southern
Colombia,
a
sequence
of
early
Neoproterozoic
vulcanosedimentary rocks of the Piraparana Formation is found covering the
Mesoproterozoic basement where an interbeded rhyodacitic flow has been dated by RbSr at 920 Ma.
5
Despite the lack of direct geochronological evidence, it has been assumed that the Paleoto early Mesoproterozoic provinces of the Amazon craton extend beneath the north
Andean foreland basins until they meet the Andes (Figures 1 and 2), where they are
thought to be truncated by exotic terranes of “Grenvillean-affinity” accreted to the
continental margin during inferred collisional interactions with Laurentia (Forero-Suarez
1990; Gómez et al. 2007; Kroonenberg, 1982).
3.2. North Andean inliers:
Along the Andean orogen, isolated inliers of Proterozoic gneisses and granulites are long
known in Colombia and they have been the focus of intensive research (MacDonald and
Hurley 1969; Tschanz et al. 1974; Kroonenberg 1982; Priem et al. 1989; Restrepo-Pace
et al. 1997; Ruiz et al., 1999; Ordonez-Carmona et al. 1999; Ordonez-Carmona et al.
2002; Cordani et al. 2005; Cardona-Molina et al. 2006; Cardona et al. 2010, among
others). However, as shown in a recent compilation of the available isotopic data
presented by Ordonez-Carmona et al. (2006), it is evident that precise geochronological
data for many of the exposed units is still lacking.
Recent paleomagnetic data acquired in the Colombian and Venezuelan Andes has shown
the strong fragmentary character of the pre-Mesozoic basement that underlies the Eastern
and Central Cordilleras of Colombia, the Santa Marta massif, and the Merida Andes and
Perija Range of Venezuela (Bayona et al. 2010; Bayona et al., 2006). It was shown by
these authors that different fault-bounded blocks in the Cordillera, some of them
associated with Proterozoic basement inliers, have suffered significant trench-parallel
latitudinal transport of differential magnitudes at least since Jurassic times. These data
most likely preclude the idea of the “Garzon-Santa Marta granulite belt” as a coherent
Precambrian terrane constituting the basement of the Eastern Andes of Colombia
(Kroonenberg 1982; Restrepo-Pace et al. 1997). Some of these blocks, as it’s the case of
the Santa Marta Massif, show displacements in excess of 100Km in magnitude with
respect to a dynamic South America reference framework (see Figure 7 in Bayona et al.
2010) while others, as the Garzón massif, are considered as being relatively stable with
respect to non-remobilized South America at least since Paleozoic times (Cardona et al.
6
2010; Toussaint, 1993). The cordilleran inliers sampled for the present study, the Garzón,
La Macarena and Las Minas massifs, represent the southernmost inliers found in the
Colombian Andes, located at approximately the same modern latitude as the wells
sampled from the Putumayo basin.
3.2.1. Garzón Massif:
This basement inlier located in the southern portion of the Eastern Cordillera of
Colombia (Figures 1 and 2) constitutes the largest exposure of Precambrian crust in the
northern Andes. Based on regional mapping and mesoscopic characteristics of the
different metamorphic rocks, it has been informally subdivided in four lithostratigraphic
units shown in Figure 2b (Jimenez Mejia et al., 2006): 1) Las Margaritas migmatites, a
metasedimentary unit located along the eastern side of the massif dominated by
metasedimentary gneisses and migmatites with mineral assemblages indicative of
amphibolite-facies metamorphic conditions; 2) El Vergel granulites, a unit dominantly
composed by felsic granulites and garnetiferous paragneisses with subordinate mafic
granulites. This unit covers most of the western flank of the massif and is in contact with
the Las Margaritas migmatites along the east-verging Las Hermosas reverse fault (Figure
2b); 3) El Recreo gneiss, an orthogneissic unit of interpreted anatectic origin, and 4) the
Guapotón-Mancagua orthogneiss, a biotite-amphibole gneiss of granitic composition that
outcrops along the western margin of the massif and displays a foliation that is
concordant with that of the El Vergel granulites. In some literature the first three units are
also referred to as the “Garzon Group” (Kroonenberg, 1982; Restrepo-Pace et al. 1997).
Geochronological data available for this massif comes from the early works of Alvarez
and Cordani (1980) and Priem et al. (1989), followed by Restrepo-Pace et al. (1997) and
Cordani et al. (2005), the last one being the most comprehensive so far and containing the
first U-Pb SHRIMP data obtained from the Precambrian of the northern Andes. These
last authors report an age of 1015±7.8 Ma for prismatic zircons extracted from a
discordant leucosome of the Las Margaritas migmatites unit, which they interpret as an
anatectic event caused by decompression melting; for the Guapoton-Mancagua
orthogneiss they obtained an age of 1000±25 Ma from metamorphic zircon overgrowths
(5 analyses) and an age of 1158±22 Ma for the magmatic precursor (3 analyses); and
7
finally, for the El Vergel granulites, zircons analyzed resulted in a complex concordia
plot from which a reliable age of metamorphism could not be extracted due to the mixed
effects of inheritance and superposed Pb-loss. With the exception of samples disturbed by
Phanerozoic thermal events, most of the Ar-Ar results from this massif cluster around
970 to 910 Ma (Restrepo-Pace et al., 1997; Cordani et al., 2005).
3.2.2. Serrania de la Macarena uplift:
The geology of the Macarena range (Fig 2) was originally studied by Augusto Gansser
(in Trumpy 1943), who described its basement as an association of mica-schists and
alkali-feldspar gneisses, hornblende gneisses, amphibolites and highly deformed
syenogranitic gneiss (Trumpy 1943, pp 1282). The Precambrian nature of these rocks was
established based on stratigraphic relations with the Güejar group, a clastic sedimentary
unit composed of micaceous and qz-rich sandstones which rests unconformably on the
metamorphic basement and contains trilobites determined as Cambro-Ordovician in age
(Harrington and Kay, 1951). However, previous to this study, no radiometric ages had
been obtained for the crystalline basement of this range. Toussaint (1993) suggested that
the basement of this sierra could potentially be correlated with the Garzon massif as part
of a larger crustal block accreted to Amazonia during Precambrian times (the so-called
Andaqui terrane), but this hypothesis is based only on stratigraphic observations that
correlate the sedimentary sequence of the Güejar series with Lower Ordovician units of
the Ariari limestone and the Ordovicico del Rio Ambica that rest unconformably on
metamorphic rocks of the Garzon group (Trumpy, 1943).
3.2.3 Las Minas Massif:
The Serrania de Las Minas is a NE trending elongated range of about 17 Km in length
and 6 Km in width, located only 20 Km west of the Garzon Massif (Figure 2b).
Structurally, this range is located along the axis of a pop-up structure bounded on both
sides by reverse faults that thrust the Las Minas migmatites and gneisses onto Tertiary
alluvial deposits of the Magdallena valley to the east, and over Jurassic
volcanosedimentary sequences of the Saldaña Formation to the west (Figure 2b). The
stratigraphic succession in this range comprises a high-grade metamorphic basement
8
overlain by clastic sedimentary sequences of Graptolite-bearing Ordovician shales and
sandstones of the El Higado Formation (Mojica et al. 1987). In some locations, this
sequence appears intruded by granitic plutons of Jurassic age (Cardona et al.
unpublished). The only geochronological analysis available for the metamorphic rocks of
this area is an hornblende Ar-Ar age obtained from an amphibolite (Restrepo-Pace et al.
1997, sample HP-3) which defines a plateau cooling age of 911±2 Ma similar to other KAr and Ar-Ar cooling ages reported for the Garzon massif (Priem et al. 1989; RestrepoPace et al. 1997; Cordani et al., 2005).
From transects performed during the present study, the metamorphic sequence was
identified as a series of stromatic migmatites overlying a metaigneous unit characterized
by porphyroclastic augen-gneisses of granitic composition. Most of the observed
migmatites are stromatic metatexites with apparent low degrees of partial melting, whose
metasedimentary origin is evidenced by relict sedimentary structures such as crossbedding observed in psammitic resistant layers that did not undergo anatexis. The
disposition of the granitic leucosomes, parallel to subparallel with respect to the dominant
metamorphic foliation in the gneisses and the overall morphology of the migmatites,
suggests that shearing and deformation occurred in simultaneous with the melting event
(e.g. Kemp and Gray 1999). Foliations measured in the gneisses and the migmatites
reveals folding about an axis oriented at ~220/10 (Figure 2b).
4. Analytical Methods: U/Pb Zircon Geochronology by LA-MC-ICP-MS
Zircon separates from drilling-core and outcrop samples were obtained by standard
methods combining jaw crushing, roller-mill pulverizing, water-table concentration,
magnetic separation using a Frantz® LB-1 barrier separator, and final heavy-liquid
separation using 3,32g/cm3 Methylene Iodide. For zircons extracted from igneous and
metamorphic rocks, approximately 50 to 100 individual grains were handpicked,
mounted in epoxy, polished, carbon-coated and imaged with SEM cathodoluminescence
(CL) prior to analysis. Zircons from sedimentary / metasedimentary samples to be
analyzed for detrital zircon (DZ) provenance were mounted randomly following the
procedures of Gehrels et al. (2006). All U-Th/Pb isotopic measurements were performed
9
by LA-MC-ICP-MS at the LASERCHRON center facilities in the University of Arizona
following the procedures presented by Gehrels et al. (2008) for analyses obtained using
the GV-Instruments Isoprobe MC-ICP-MS. However, most of the analyses were obtained
using a NU-instruments HR-MC-ICP-MS coupled to a New Wave UP193HE Excimer
laser operating at a wavelength of 193 nm. Details about the analytical procedures
followed are described in the appendix. Both multicollector blocks are capable of
simultaneously measuring
while
238
U,
232
Th,
208
Pb,
207
Pb, and
206
Pb isotopes in faraday cups
204
Pb is always measured in a channeltron® detector or an ion-multiplier. This
configuration was utilized for analyses performed using ablation spot-sizes between
40µm and 22µm in diameter. Alternatively, in very complexly zoned zircons, we utilized
10µm or 18µm laser spot-sizes in which case masses 208, 207, 206 and 204 were all
measured in ion-multipliers in order to allow reliable measurement of the Pb isotopes
with low ablation volumes, following procedures similar to those presented by Johnston
et al. (2009).
All errors reported in the data tables are analytical only. Systematic errors for the
measurement of the
206
Pb/207Pb ratios on the Sri Lanka (SL2) standard varied for each
analytical session, but were normally between 0.8 and 1.2 %. Final ages reported in the
concordia plots and discussed throughout this paper represent 2σ confidence-limit total
errors, and account for the quadratic propagation of both the analytical and systematic
uncertainties. Concordia plots, age calculations and probability density diagrams were
plotted at a 1σ uncertainty using the Isoplot Excel® macro of Ludwig (2003).
For metasedimentary samples, analyses of inherited cores were performed carefully using
the CL images in order to avoid areas of obvious metamorphic recrystallization.
However, as will be evident from the concordia plots below, some of the analyses show
departures from concordia along a discordia chord that goes towards the age of the
metamorphic overprint while others are apparently affected by normal (young) Pb-loss
events. Knowing this, caution must be taken when interpreting one-dimensional
probability density plots that consider only one isotopic ratio for the calculation of the
best age (i.e.
206
Pb/207Pb in this case), as Pb-loss events at non-zero ages can generate
10
“false” distribution peaks and lead to misinterpreting detrital zircon signatures (as
illustrated by Nemchin and Cawood, 2005). For the probability density plot of each
sample presented here (Figure 8), we took an approach similar to Collins et al. (2007) and
plotted both the entire dataset of each sample (shaded in red), and the data filtered at <5%
discordance (black solid line). Although some of the smaller peaks that are present in the
plots of the entire dataset disappear when the data is filtered, the position of the dominant
peaks and their correspondent age maxima do not show any significant variation, which
gives further confidence in the robustness of the obtained DZ data.
5. Samples description and geochronological results:
For the present study we analyzed samples of a) gneisses, granulites and migmatites
collected from three Cordilleran Inliers of the Northern Andes (the Garzón, Macarena
and Las Minas massifs); b) drilling-core samples of deep exploratory wells that pierced
the crystalline basement of the Putumayo foreland basin of Colombia, east of the frontal
thrust systems of the northern Andes, and c) undeformed or slightly deformed intrusive
rocks from the northwesternmost cratonic exposures in the Amazonas region of
Colombia. Sample locations and other geographic names referenced in the text are shown
in Figures 1 and 2.
5.1. Cordilleran Inliers:
Seven samples collected in three of the basement inliers were analyzed (Figure 2), three
of them from the Garzón massif (MIVS-11, MIVS-26 and CB-002), one from the
Serrania de la Macarena range (Macarena-2), and the other three from Las Minas massif
(MIVS-37a and MIVS-41 and CB-006). One sample from the La Plata orthogneiss in the
Central Cordillera, a metamorphic unit of alleged Precambrian age (Gómez et al. 2007;
Priem et al. 1989) was also analyzed but given the Upper Permian age of crystallization
obtained it will not be discussed further (Ibanez-Mejia, unpublished).
5.1.1. Garzón Massif:
Sample MIVS-11 was collected from the El Vergel granulites unit, approximately 7 Km
southeast of the Garzón municipality along the road that connects Garzón with the town
11
of El Recreo (Figure 2b). It is a hyperstene-hornblende-K feldspar-plagioclase-quartz
felsic granulite, with slight amphibolite-facies retrograde overprint. Zircons recovered
from this sample were mostly rounded to subrounded in shape, colorless to light pink,
with 1:1 to 1:3 width:length ratios and generally smaller than 300µm in size. CL imaging
shows that most of them exhibit textures indicative of growth and/or recrystallization
under metamorphic conditions (inset, Figure 3a) evidenced by relatively homogeneous
grains displaying only weak sector-zonation (Corfu et al. 2003), or rims with no internal
structure that appear as overgrowths around more complexly zoned cores. U-Pb isotopic
analyses of the metamorphic overgrowths are all concordant (Figure 3a), and yield an age
of 992±10 Ma (MSWD=0.32, 27 analyses) that is interpreted as the timing of granulitefacies metamorphism for this unit. Uranium concentrations range between 200 and 600
ppm with fairly low U/Th ratios between 1 and 2. A 206Pb/207Pb weighted mean age of the
same 27 grains results in an age of 995±10 Ma that is in excellent agreement with the
concordia age. As described above, many of the grains contain xenocrystic cores, some of
which are characterized by an oscillatory zoning typical of magmatic zircons while others
exhibit complex chaotic zoning patterns. The wide range of ages obtained for these
xenocrysts, ranging from ca. 1011 to 1490 Ma (Repository Data), reflects the detrital
nature of these grains inherited from a volcanosedimentary protolith. A total of 89 detrital
core analyses were performed, 62 of which are <5% discordant. The pattern is
characterized by a multi-modal age peak distribution dominated exclusively by
Mesoproterozoic grains, with the most prominent maxima at ca. 1120, 1300, 1370 and
1450 Ma. A maximum age of deposition for the sedimentary protolith is estimated from
the two youngest, >95% concordant grains with ages of 1115±10 and 1118±22 Ma
respectively (1σ, total error). This places a limit for the age of sedimentation that is thus
constrained between the age of the youngest grain (1115±10 Ma), and the age of
metamorphism (992±10 Ma).
Sample MIVS-26 was collected in a small quarry close to the Guapotón municipality
(Figure 2b). It corresponds to an augen orthogneiss with centimetric K-feldspar
porphyroclasts in a matrix of finer-grained quartz, plagioclase, biotite and hornblende
aligned defining the metamorphic foliation. Although no quantitative P-T estimates of
12
metamorphism are available for this unit, the mineral assemblage suggests amphibolitefacies conditions (Jimenez Mejia et al. 2006; Kroonenberg, 1982). No evidence for
partial melting was observed in this outcrop. Zircons from this sample are mostly
uncolored, ranging in size from ~150 to 300µm, with width:length ratios of 1:2 to 1:3. CL
imaging of 50 handpicked grains shows that they consist of oscillatory-zoned igneous
cores rimmed by more homogeneous, unzoned, metamorphic overgrowths (inset, Figure
3b). All of the concordant U-Pb analyses from the igneous cores yield an age of 1136±11
Ma with an MSWD=0.84 (Figure 3b, 24 analyses less than 5% discordant). This age is
thought to represent the time of crystallization for the igneous protolith, and is within
error of the SHRIMP age of 1158±22Ma reported by Cordani et al. (2005) for this same
unit. Analyses performed on the rims result in an age of 990±12 Ma (MSWD=1.15),
calculated from 15 rim analyses that are less than 5% discordant. We interpret this age to
reflect the timing of peak metamorphism of this metagranite which is synchronous with
granulitization of the El Vergel unit (obtained from samples MIVS-11 and CB-002), and
slightly younger but more precise (although within error) than the 1000±25Ma age
reported by Cordani et al. (2005). Nine analyses were not considered for the calculation
of either the igneous or metamorphic age, as they possibly represent mixtures between
the core and the rim during ablation or partial Pb-loss by recrystallization. These analyses
are shown in light gray in Figure 3b.
Sample CB-002 is a leucocratic gneiss also collected from the El Vergel granulites unit,
along the road that connects the towns of Garzón and Guadalupe (Figure 2b). This
outcrop consists of sillimanite-bearing quartz-feldspar gneisses with biotite as the only
ferromagnesian bearing phase also interpret to be of metasedimentary origin. Zircons
from this sample are mostly uncolored, but relatively more prismatic than those of sample
MIVS11 displaying width:length ratios from 1:3 to 1:5. CL imaging shows a wide variety
of internal grain morphologies similar to sample MIVS11, although elongated zircons
with igneous zoning are more abundant. This sample shows evidence of extensive solidstate recrystallization of inherited zircon cores by the granulite-facies metamorphic event,
observed as widespread “ghost” oscillatory zoning patches in recrystallized portions of
zircon grains and the presence of recrystallization fronts (e.g. Hoskin and Black, 2000).
13
This feature is also evident in the U-Pb systematics, where ages obtained from these
recrystallized areas show some “memory” of their original age component, and are
slightly shifted away from the age of metamorphism towards older ages. The analysis of
10 metamorphic overgrowths and individual crystals that display homogenous, bright and
unzoned textures under CL, are all concordant and consistent with an age of 992±10 Ma
(MSWD=0.3) for the metamorphism. This age is equivalent within error to the
metamorphic ages obtained from the other two samples of the Garzón massif described
above. For the analysis of the detrital component, given the strong metamorphic overprint
observed in the zircon cores, a more cautious approach was taken for this sample.
Analytical spots where carefully selected away from areas where recrystallization is
evident and placed in cores where pre-existent zoning is better preserved. Additionally,
results plotted in the probability plot if Figure 8 were filtered to discard xenocryst grains
whose calculated
206
Pb/207Pb age overlaps within analytical error with the
abovementioned age of metamorphism in an attempt to reduce the “noise” introduced by
recrystallization. All analytical data, including that of those grains that were excluded
from plotting, are reported in the repository data table. The detrital age spectrum for this
sample shows a multi-modal age peak distribution of early Mesoproterozoic components
similar to sample MIVS-11, with peak modes at ca. 1340, 1420, 1450 and 1480 Ma
(Figure 8). Considering that young detrital cores whose ages overlap within error with the
age of metamorphism were discarded, the maximum depositional age obtained with the
remaining grains is a conservative estimate for the maximum age of sedimentation. The
youngest four, <5% discordant grains, have ages of 1011±10, 1013±10, 1014±11 and
1014±11 (1σ, total error), respectively. This gives us confidence to place the age of
deposition for the sedimentary protolith between 1013±11 (Weighted mean of the 4
youngest analyses) and the age of metamorphism at 992±10 Ma.
5.1.2. Serrania de la Macarena uplift:
One sample was analyzed from the metamorphic basement of this sierra (Macarena-2). It
consists of a melanocratic biotite-epidote-microcline-plagioclase-qz mylonitic gneiss
with abundant sphene, suggestive of dynamic metamorphism under at least uppergreenschist-facies conditions. Sub-grain rotation deformation mechanisms observed in qz
14
and feldspar crystals suggest intermediate temperatures for the mylonitization event
possibly within the 400-600°C temperature range, but that did not generate a strong
metamorphic imprint on U-Pb systematics of the analyzed zircon crystals (Figure 4). CL
images from handpicked zircons show that most of the grains have oscillatory-zoned
textures indicative of igneous crystallization (inset, Figure 4). Most of the point analyses
performed were >95% concordant, and a concordia age calculated with 19 individual
analyses yields a value of 1461±15 Ma (MSWD=0.37) for the igneous precursor of these
mylonites. In some grains, the normal igneous textures appear to be partially obliterated
by partial recrystallization of the zircons (possibly during the mylonitization event), but
spot analyses performed in these areas resulted in only slightly discordant ages that are
very close to those obtained from the unaffected portions of the grains. This age of 1.46
Ga, along with a protolith age of ca. 1.5 Ga reported for an orthogneiss of the Serrania de
San Lucas massif (Ordonez-Carmona et al. 2009), are the oldest protolith ages known for
metaigneous units of the north Andean inliers to date. No units of similar age have been
found in the Garzon massif, where zircons of this age are only found as detrital
components in the metasedimentary units presented in this paper. Although this age does
not conclusively confirm or preclude correlations that suggest the basement of this sierra
to be part of the Andaqui terrane along with the Garzon massif (Ramos, 2010; Toussaint,
1993), the protolith of these mylonites and correlative units may have served as a source
for the ca. 1.45 Ga detritus found in metasedimentary units of the Garzon group. It should
also be noted that igneous or metamorphic rocks of similar age have also not been found
in either cores or outcrops of the stable cratonic area or basement cores as will be shown
below.
5.1.3. Las Minas Massif:
Three samples were analyzed from the Minas massif: MIVS-41, collected from the
granitic augen-gneiss that forms the core of the massif; CB-006, a biotite-rich
melanosome from a metasedimentary migmatite in proximity to the orthogneissic unit;
and MIVS-37a, collected from an outcrop of metasedimentary qz-feldspar migmatitic
gneisses on the northern termination of the range (Figure 2b).
15
Sample MIVS-41 is a hornblende-biotite-microcline-plagioclase-quartz augen-gneiss that
forms the core of the massif and is well exposed along the Zancudo creek that drains the
eastern portion of the Sierra. Along this creek, outcrops of granitic gneiss are apparently
overlain by metasedimentary migmatites although the contact between the two units was
not observed. Zircons from this sample are light pink in color, have width:length ratios
between 1:3 and 1:4, and reach up to 400µm in length. Internal grain structures are
dominated by igneous oscillatory-zoning, which in some grains is affected by partial
recrystallization caused by the amphibolite-facies metamorphic event (Figure 3d). U-Pb
analyses performed on the igneous cores reveal a mid-Mesoproterozoic age for the
magmatic precursor of the gneiss, calculated at 1326±13 Ma using 23 analyses that are
less than 5% discordant. There is a subset of analyses that, although <5% discordant,
clearly define a mixing chord between the age of the igneous precursor and the age of
metamorphism, demonstrating that the recrystallization mantles that surround the igneous
grains are zones where only partial resetting of the U-Pb systematics took place. This
clearly exemplifies how individual spot ages that lie along this chord, although
overlapping the concordia within error, lack any geological meaning by themselves as
they represent mixing between two end age-components and not a real tectonothermal
episode. Using the ion-counter configuration we were able to obtain a group of
concordant analyses from thin, unzoned, brighter metamorphic overgrowths, which yield
an age of 991±12 Ma for the metamorphic event (8 analyses, MSWD=1.08). This age is
basically the same within error with respect to the age of granulite-facies metamorphism
previously shown from samples of the Garzon massif. Granitic protolith ages of ca. 1.3
Ga have not been found anywhere else in the Andean inliers, but zircons of this age are
widespread in the many of the metasedimentary samples discussed throughout this paper
(Figure 8).
Sample CB-006 was also taken along the Zancudo creek, a few tens of meters
downstream (east) from the location where the MIVS-41 sample was collected. In this
location, metatexic migmatites of metasedimentary origin display well developed
stromatic structures, characterized by an interleaving of 5 to 10 cm thick granitic
leucosomes with laterally persistent bands of biotite-hornblende gneisses. The sample
16
analyzed corresponds to a sillimanite-bearing biotite rich gneiss with amphibole and
clinopyroxene interpreted to be a melanosome for these migmatite. Zircons recovered are
mostly rounded to subrounded in shape, displaying width:length ratios of 1:2 to 1:3 and
average sizes between 50 and 250 µm and CL imaging showed that most of the grains
have xenocrystic cores surrounded by bright, unzoned overgrowths presumably
developed during metamorphism and partial melting. Interestingly in this sample, the
majority of the xenocrysts are characterized by textures indicative of growth under
metamorphic conditions (sector-zoning, inset Figure 3e), and low-U concentrations
typically between 70 and 250 ppm (see supplementary data). This observation contrasts
with all the other metasedimentary samples presented in this study, where most of the
xenocrystic cores display textures indicative of igneous conditions of crystallization. The
probability age plot for the detrital component shows an almost unimodal age
distribution, dominated by a restricted population within the range from 1100 to 1200 Ma
(Figure 8). Grains that fall outside of this range typically have magmatic zoning, but
some of them are also low in uranium. This is the case of the youngest detrital grain
analyzed which, with an U content of only 83 ppm, results in an rather imprecise age of
1023±81. A weighted mean age calculated using the youngest 4 grains constrains the
maximum timing of deposition for the sedimentary protolith to 1086±34 (2-sigma).
Sample MIVS-37a was collected from the northwestern termination of basement
exposures in the Las Minas massif. It consists of a leucocratic, quartz-feldespar
migmatitic gneiss with biotite (extensively retrograded to chlorite) as the only
ferromagnesian metamorphic phase. The outcrop is crosscut by a series of slightly folded
mafic dikes of unknown age. The processed sample is a paleosome from these migmatites
in a portion where leucocratic veins are scarce, selected in order to obtain inherited
zircons that represent the detrital provenance of the sedimentary precursor. CL imaging
showed that most of the grains have igneous growth-zoning patterns and resorption
textures, but lack evident metamorphic overgrowths or extensive recrystallization. U-Pb
analyses confirm that the zircon grains are detrital in character and that zircon did not
recrystallized during the metamorphic event but, instead, was probably resorbed during
anatexis as a result of high-temperatures and Zr subsaturation in the partial melts. The
17
age spectra obtained, considering all the analyzed grains, ranges from 1005±23 Ma to
1516±21 Ma. The dominant distribution range has a peak maximum at ca. 1490 Ma,
while grains that fall out of this range constitute small subsidiary peaks at ~1160 and
~1210. The youngest grain analyzed, slightly more than 5% discordant, has a 206Pb/207Pb
calculated age of 1005±23 Ma. However, maximum depositional ages inferred from only
one analysis that does not form part of a grain population are not always reliable. For this
reason, we adopt an age of 1157±27 as a conservative estimate of maximum protolith
deposition, calculated as the weighted mean of the youngest, <5% discordant, two grains
of the distribution that form part of a coherent population.
5.2. The Basement underlying the Foreland Basins:
The North Andean foreland basins of Colombia can broadly be divided in two sub-basin,
namely the Putumayo to the south which represents the northern extension of the
Marañon-Oriente basins of Peru and Ecuador, and the Llanos to the north that continues
northeastward into Venezuelan territory (Figures 1 and 2). These two basins are separated
by a series of NW trending basement structures, the Macarena-Chiribiquete structural
high and the Vaupes arch, which define the elongated Vaupes-Amazonas intracratonic
basin extending from the borders with Peru and Brasil in southeastern Colombia to the
Serrania de la Macarena uplift next to the Andean chain. These basement–arch structures,
roughly located along the southern and northern margins of the Güejar-Apaporis Graben
(Figure 2), are readily observed in the Bouger gravimetric anomalies map of Colombia
(available free for download at the Colombian National Hydrocarbons Agency webpage,
www.anh.gov.co). This structure has been interpreted as the expression of a lower
Paleozoic graben that trends NW-SE and controlled the sedimentation of the Lower
Paleozoic Araracuara Formation (Cáceres et al. 2003).
Many exploratory wells in the foreland basins of Colombia have drilled below the
unconformity between the crystalline basement and the overlying Phanerozoic
sedimentary cover, but unfortunately not many of them have retrieved core samples. The
depth to the crystalline basement is highly variable, depending on the relative position of
each well with respect to the horst/graben basement structures. However, the crystalline
18
basement is generally shallower in the Putumayo basin in comparison to the Llanos. In
the Putumayo basin, on those wells where basement cores have been recovered,
Cretaceous rocks directly overlie the Precambrian units whereas a pile in excess of 1Km
in thickness has been drilled in several locations around the Llanos basin. For this reason,
it is more common to find wells piercing the crystalline basement and recovering cores in
the Putumayo area than in the Llanos.
For the purposes of this study, we sampled six different exploratory wells that recovered
core samples from the crystalline basement of the Putumayo Basin. From north to south
they comprise the Payara-1, La Rastra-1, Solita-1, Mandur-2, Caiman-3, and Mandur-3
wells (Figure 2), which drilled through basement rocks below depths of 3492’ft, 3087’ft,
3675’ft, 5613’ft, 7704’ft and 7508’ft, respectively. These are not the only wells that
reached the crystalline basement in the region but, to our knowledge, are most of those (if
not the only ones) that recovered cores that are accessible for sampling. These cores
represent the only available means to study the petrology and geochronology of the
basement of the proximal Andean foreland in Colombia. Samples processed from the
Mandur-3 and La Rastra-1 wells did not yield zircons concentrates so they will not be
discussed further. They are an epidote-amphibolite -facies metabasite rock, and a kyanitebearing grt-cpx-amph-bt-plag-qz melanocratic schist. These two samples are being
analyzed by other methods and will be presented in a future paper.
5.2.1. Putumayo Basin Basement Cores:
The Payara-1 well was drilled in the northern part of the basin and is the northernmost of
the basement wells presented in this study (Figure 2a). It drilled through a 3492’ft thick
Meso- Cenozoic sedimentary succession to finally pierce the crystalline basement along
the western flank of the Macarena arch. The core sample consists of a granulite-facies
Opx-Sil-Grt-Amph-Bt-Plag-Qz migmatitic gneiss (Figure 5a), and the occurrence of
feldspar-quartz symplectites intergrown with fleaky biotite is suggestive of reactions
involving biotite dehydration melting (Waters, 2001). CL images of zircons recovered
from this sample all have oscillatory-zoned cores that indicate formation under igneous
conditions, which in turn appear mantled by a thin rim of unzoned bright metamorphic
19
zircon (inset, Figure 6a). Small zircons grains tend to show a subtle fading of the original
igneous growth zoning which can be due to the effect of partial solid-state metamorphic
recrystallization under granulite-facies conditions (e.g. Connelly, 2001). This observation
is supported by the U-Pb isotopic data, where it is evident that spot analyses performed in
smaller grains are slightly discordant and lying along a chord that joins the bigger, well
zoned, and isotopically concordant cores in the upper discordia intercept with the age of
the metamorphic overgrowths in the lower intercept (Figure 6a). Calculation of a
concordia age using 16 core analyses that are <3% discordant yields a value of 1606±16
Ma (MSWD=0.97) for the crystallization of the igneous protolith. Given the thin width of
the metamorphic overgrowths (<30µm), analyses were performed using an 18µm
diameter spot size with the ion-counter configuration. Seven analyses were acquired from
these overgrowths, four of them being <4% discordant. One is 10% normal discordant,
another one is 8% reverse discordant, and the last one is 5% discordant but is
significantly older than the others and lying along a chord that goes towards the age of
the igneous cores. We interpret this last analysis as a mixed ablation spot. The four <4%
discordant analyses yield an age of 986±16 Ma (MSWD=0.49) for the granulite-facies
event, considered to be the best estimate for the age of metamorphism of this gneiss. A
206
Pb/207Pb weighted mean age of the 6 non-mixed analyses defines an equivalent but
less-precise age of 985±23 Ma (MSWD=0.70), but as in previous samples the concordia
age is preferred. Note the very low U concentrations measured on the overgrowths,
estimated between 15 and 20 ppm (supplementary data). This result demonstrates the
occurrence
of
Grenville-age
granulite-facies
reworking
of
Amazonia’s
late
Paleoproterozoic crust, just east of the Garzon massif in the proximal foreland basement.
The Solita-1 well was drilled in the central part of the basin, coring basement rocks of the
Florencia arch below a 3675’ft thick sequence of Cenozoic clastic rocks. The analyzed
sample consists of strongly deformed Amph-Bt-Ep-Plag-Qz migmatitic gneisses, which
suggests syn-deformational metamorphism and melting under amphibolite facies
conditions. The folded leucosomes and the foliation of these migmatites appear crosscut
by discrete blastomylonitic zones along which euhedral amphiboles and subhedral
pyroxenes attain larger grain sizes than the regional metamorphic phases of the
20
migmatite. It is unclear whether if the mylonitization event is somehow related to the
phase of regional metamorphism and anatexis (slightly postdating it) or if it is a
considerably younger event that was not recorded in the zircon U-Pb system of the nonmylonitized rock. Recovered zircons are small in size, generally <150µm in length,
displaying 1:2 to 1:4 width:length ratios. Most of the crystals are cloudy with reddish and
brownish colors and show strong suppression of their CL emission, observations that
suggest a considerable degree of metamictization (inset, Fig 6a). Xenocrystic cores are
common in most of the imaged crystals and U-Pb analyses performed on zircons from
this sample produce a quite complex concordia plot. Most of the spots placed on the dark
overgrowths appear to be aligned defining a discordia chord with an upper intercept of
1044±26 Ma and a lower intercept trending towards recent Pb-loss (Figure 6b). This
upper intercept, considered to represent the age of metamorphism and melting, is
calculated using 12 analyses that are less than 30% discordant, and anchoring the lower
intercept at 6 Ma. Two important geological events that affected the western Putumayo
basin justify the lower-intercept anchoring of the discordia to late Miocene times, as they
may be responsible for the observed Pb loss: 1) Hydrothermal water circulation from the
Cordillera towards the foreland during Oligo-Miocene Andean uplift pulses and orogenic
development (Parra et al. 2009), and 2) thermal effects and hydrothermal water
circulation during nearby late Miocene mafic magmatism. In the Orito oil field, gabbroic
dikes are known to cut the Cretaceous sequence of the southwestern Putumayo basin
where they have been dated at 6.1±0.7 Ma by Amphibole K-Ar (Vasquez et al. 2009). A
weighted average age calculated using 11 analyses that are <20% discordant produces an
undistinguishable age of 1046±25 Ma for the migmatization event. Ellipses shown in
blue in Figure 6b represent analyses that are either xenocrystic cores or ablation spots that
represent mixtures between the inherited component and the ca. 1.05 Ga overgrowths.
Older ages, with ages of 1.73 and 1.85 Ga, are interpreted as detrital grains from a
sedimentary protolith. Some of the mixed analyses appear to be aligned along a chord
that joins the metamorphic age and a presumable ca. 2.0 Ga component.
The Mandur-2 well cored a depth interval between 5690’ and 5708’ beneath the westcentral part of the basin, and consists of an intercalation of melanocratic migmatitic
21
gneisses and centimetric to decimetric coarse-grained leucocratic veins and dikes of
granitic composition (Figure 5b). Given the size of the granitic layers and the lack of
evident melanosomes and/or reaction margins (selvedges) along the contact with the
surrounding gneiss, we interpret these dikes as granitic melts injected into the
melanocratic gneisses within a regional migmatitic terrain, but bearing no genetic relation
(as derivation through partial melting) with its more mafic gneissic host rock. The
granites display clear textures of crystallization from a silicate melt but also display flow
and shearing textures indicative of injection and crystallization under differential stress
conditions. These dikes are petrographically classified as syenogranites, composed
mostly by microcline and quartz with subordinate plagioclase and muscovite. The
melanocratic host rocks for the dikes are Amph-Bt-Plag-Qz gneisses that also underwent
some degree of partial melting during the metamorphic peak at least under amphibolite
facies conditions. Thin leucocratic stromata were developed parallel to the foliation of the
gneisses, and their white color contrasts with the pinkish tone of the syenogranitic dikes.
Additionally, these elongated patches of partial melt (neosome) are trondjhemitic in
composition, completely lacking any K-feldspar in contraposition to the microcline-rich
granitic dikes. We thus interpret the overall stromatic morphology as the result of layerparallel injection of the syenogranitic melt, although the presence of thin trondhjemitic
neosomes in the gneisses also evidences some degree of partial melting of the host rocks.
We analyzed two separate samples from this core; one zircon concentrate was obtained
from a sample of an approximately 70 cm thick coarse-grained leucocratic dike, and the
other from the host melanocratic amphibole-gneisses. Zircons from the gneiss display
complex internal morphologies under CL (inset, Figure 6c), characterized by cores with
oscillatory zoning patterns that are enclosed by low-luminescent, high-U rims inferred to
have crystallized under metamorphic conditions. Textures indicative of solid-state
recrystallization were also observed, evidenced as bright margins with lobate
terminations inboard of the metamorphic overgrowths that clearly crosscut the primary
igneous zonation of the cores. These rims resemble the recrystallization fronts studied by
Hoskin and Black (2000). Isotopic analyses performed on the low-luminiscent
metamorphic overgrowths are concordant to moderately discordant (Figure 6c), yielding
22
an upper discordia intercept age of 1019±23 Ma (MSWD=0.62) for the amphibolitesfacies event similar to that found in the Solita-1 well but older than the granulite-facies
imprint dated from the Payara-1 well. Analyses of the xenocryst cores are more variable,
depending primordially on the internal structure of each grain. Most of the cores with
oscillatory-zoned textures appear to pertain to a single population variably affected by
Pb-loss effects, defining a discordia chord with a latest Paleoproterozoic upper-intercept
age of 1601±28 Ma (MSWD=0.62). Cores with non-oscillatory zoned textures yield older
ages up to 1713 Ma, which can be inherited from a vulcanosedimentary protolith. Finally,
analyses performed on the bright luminescent areas showed mixed ages between the
inherited components and those of the metamorphic rims (gray ellipses in Figure 6c),
interpreted as mixed ablation spots and/or partial metamorphic recrystallization of
inherited zircon cores.
Zircons from the leucocratic sills are mostly elongated and prismatic, displaying
width:length ratios between 1:3 and 1:5. Some of the grains were broken during sample
processing due to their big sizes of at least 500µm in length. Internal CL images showed
that the grains have oscillatory-zoned textures indicative of crystallization from a melt,
but some internal complexities are apparent from the presence of bright rims surrounding
darker cores which sometimes appear to truncate previous zoning patterns (inset, Figure
6d). Around 60 U-Pb spots were performed over the different textural areas and
complexities of the grains but they produced a fairly simple plot (Figure 6d), with most of
the analyses overlapping concordia within error and defining a rather homogeneous
population with few analyses affected by Pb-loss (gray ellipses) and others evidencing
some degree of inheritance from an older component (blue ellipses). An homogeneous
population defined by 52 analyses that are <6% discordant produced a concordia age of
1017±10 Ma with an MSWD=0.66. This age, interpreted as reflecting the timing of
crystallization of the granitic melt, is more precise but still within error with respect to the
age of metamorphism measured from the zircons overgrowths of the host gneisses
previously described. We thus consider that regional metamorphism and injection of
granitic melt were relatively synchronous in this area of the basin (within analytical
error), and that this age of 1017±10 may also be a better estimate for the amphibolite-
23
facies metamorphism of the gneisses. These geochronological evidences further support
the textural interpretations presented above with regards to the temporal relations
between metamorphism, partial melting and injection of syenogranitic sills observed in
the core sample.
The Caiman-3 well cored approximately 30 feet of basement in the southwestern part of
the basin, sampling a depth interval between 7704’ and 7731’ just below the
unconformity with the overlying Cretaceous strata. This 27 ft long core broadly consists
of two different rock types: An upper part of felsic migmatites (Figure 5c), and a lower
part of leucogranites that crosscut the foliation of the metamorphic host rocks (Figure
5d). The migmatites display leucosomes that are sub-parallel with respect to a foliation
defined by the orientation of biotites, quartz and plagioclase, and judging by the relative
proportions of leucosome with respect to the apparent paleosome this could tentatively be
classified as a diatexite. However, as this rock was located just a couple meters below the
Cretaceous paleosurface, the entire core is intensively weathered making more detailed
petrographic observations impossible. Zircons extracted from a segment of the core
between 7718.5’ and 7719’ (corresponding to Figure 5c) have rounded to sub-rounded
external shapes and length:width ratios from 1:1 to 2:1. CL images revealed a complex
internal morphology for the imaged grains, characterized by cores with oscillatory-zoning
that are mantled by low luminescent thin metamorphic rims (inset, Fig 6e). Individual
grains are generally not larger than 110µm, and metamorphic rims do not exceed 20µm in
thickness. We performed 85 spot analyses in the xenocrysts utilizing a 30µm laser
diameter, and 20 analyses on the overgrowths using a 10µm laser-beam diameter on ioncounter configuration. From the rim analyses, only 12 clustered around an age of
~990Ma while the remaining gave ages that were much older, reflecting partial zircon
recrystallization and/or mixing of old core material in the ablation sites. Figure 6e shows
the concordia plot for this sample, where green ellipses refer to analyses performed using
the 10µm spot-size; red ellipses refer to <5% discordant analyses performed with a 30µm
diameter beam, and gray ellipses are >5% discordant analyses also with a 30µm beam. A
concordia age calculated from the rims data yields a value of 989±14Ma (MSWD=0.7),
interpreted as the age of the metamorphic event. From the xenocryst core results 11
24
grains with >30% discordance were discarded from ploting, while the remaining 74
revealed that all the zircons analyzed were derived from late Paleproterozoic to early
Mesoproterozoic source areas like those found in the northwestern Amazon Craton. From
the concordia plot of Figure 6e it can readily be observed that many of the analyzed
xenocrysts have been affected by varying degrees of Pb-loss, possibly by partial
recrystallization during the metamorphic event. If the observed discordance is in effect
related to metamorphism, individual ellipses could conceivably be displaced from their
“true” age towards slightly younger apparent 206Pb/207Pb ages if they lie along a discordia
chord that is subparalel to concordia joining the original age of each individual grain (as
the upper intercept) with the age of the metamorphic event in the lower intercept (e.g.
Nemchin and Cawood, 2005). This effect however, although difficult to quantify, would
only reinforce the observation that all of the detrital grains analyzed have crystallization
ages older than about 1.45 Ga as any “partially reset” ages would yield slightly older
upper intercepts if the Pb-loss event goes through 0.99 Ga (age of metamorphism) instead
of being projected through the origin. The diverse range of ages obtained for the
xenocrystic cores reflects the detrital nature for the protolith of this rock, providing
valuable information about the provenance of this unit. The age spectrum, which is
shown in Figure 9, is dominated by peak modes at ca. 1470 Ma, 1500 Ma, 1570Ma, and
in the range 1650-1680Ma.
The second sample analyzed from the Caiman-3 well, corresponding to the lower
segment of leucogranites, was taken from the depth interval between 7728.5’ and 7729’.
It is petrographically classified as a monzogranite with muscovite and it crosscuts the
foliation of the migmatites described above. Zircons recovered have more elongated and
prismatic shapes than those from the migmatites, display width:length rations between
1:2 and 1:4, and are in this case characterized by bright overgrowths enclosing darker
xenocryst cores with crystal terminations that tend to be sharper in comparison to the
rounded edges of the zircons from the host migmatite (inset, Figure 6f). Additionally,
some of the overgrowths were observed to have oscillatory-zoning patterns, indicating
crystallization from a melt. Analyses from this sample were carried out the same way as
in zircons from the migmatites, using 10µm ion-counter configuration for the rims and
25
30µm faraday configuration for the cores. Twenty-three spots were analyzed from the
overgrowths but only 7 of them were younger than ~1Ga, reflecting significant mixing of
older material in most of the ablation pits. Those 7 rim analyses are mostly <5%
discordant yielding a concordia age of 952±21 Ma (MSWD=0.98) for the crystallization
event, confirming the textural relationships observed from the core as the leucogranites
are in fact younger (and distinct within error) than the host felsic migmatites. This age of
crystallization is important as it places a minimum age for pervasive deformation in this
region of the basement, and is comparable to ages reported for postectonic pegmatitic
dikes in the Mexican inliers (Solari et al., 2003). From the xenocryst cores, 61 out of the
84 analyses performed are <5% discordant and they yield ages ranging from ca. 1440 to
1700 Ma with distribution peak maxima at ~1460, ~1490, ~1530, ~1615 and ~1677 Ma.
This age distribution closely resembles that of the migmatites from this same well (Figure
8), suggesting that these leucogranites may represent anatectic melts derived from
metasediments similar or related to those of the host rocks.
5.3. Cratonic exposures:
Samples analyzed from outcrop exposures of the Amazon Craton in eastern Colombia are
all undeformed or sheared granitoid rocks, analyzed in order to constrain a minimum age
of significant metamorphism and deformation within this largely unknown portion of the
shield. They are subdivided into the Araracuara granitoids, collected from a basement
window along the Yari and Caqueta Rivers in proximity to the Araracuara Indigenous
settlement, and the Mitu-Taraira granitoids, outcropping in easternmost Colombia in the
Vaupes department along the Vauper and Apaporis rivers (Fig2).
5.3.1. Araracuara Granitoids:
Sample PR-3215 is classified as a sheared biotite syenogranite. It is composed of quartz,
alkaline feldspar, plagioclase, brown biotite with occasional chloritized margins, and
sporadic epidote (seemingly pistacite). Two generations of K-feldspar are observed in the
thin section, a probably older generation of orthoclase showing evidence of strong
cataclastic deformation, while the more abundant microcline is coarser-grained and
appears less deformed. Twinned microcline megablasts are common and display
26
poikilitic textures, containing inclusions of plagioclase, quartz and less frequently biotite.
Some crystals of plagioclase included or in direct contact with the K-feldspar exhibit
myrmekitic textures, and quartz crystals appear elongated, displaying ondulose extinction
and contorted margins. These observations altogether suggest that the granitic rock has
been deformed under dynamic low- to intermediate-temperature metamorphic conditions
resulting in grain-size reduction and orienting of the quartz crystals, recrystallization of
orthoclase as microcline, and are responsible for the partial replacement of plagioclase
for K-feldspar observed as myrmekitic structures. Solid-state recrystallization may have
occurred in a temperature range between 400-450°C, constrained by the observed
deformation mechanisms of quartz and feldspars (Passchier and Trouw, 2005). U-Pb
geochronological data obtained from zircons of this sample all define a single population
with variable degrees of Pb-loss (Figure 7a), with no evidence for the presence of older
inherited material. CL images obtained from the analyzed grains (inset, Figure 7a) show
internal oscillatory zoning textures typical of zircons crystallized from a melt, and many
of the grains seem to be affected by significant metamictization. A concordia age
calculated using the 14 most concordant analyses (<2% discordant) yields a value of
1756±18Ma (MSWD=0.98), equivalent to a
206
Pb/207Pb weighted mean age of
1750±18Ma (MSWD=0.43) calculated using the entire dataset. The metamorphic
conditions that affected the original granitic rock were not of sufficient grade to induce
zircon recrystallization and/or formation of metamorphic overgrowths, and we consider
the age of 1756±18 Ma as reflecting the age of crystallization of the granitic protolith.
Sample J-263 has essentially the same mineralogy as sample PR-3215 but it is less
affected by cataclasis and recrystallization. Orthoclase is more abundant that microcline,
and other deformational features like myrmekites and grain-size reduction are less
widespread. Original igneous textures such as dominantly euhedral plagioclase
surrounded by anhedral quartz grains filling spaces are better preserved, evidencing their
igneous origin (Vernon, 2004). Zircons from this sample also show oscillatory zoning
under CL (inset, Figure 7b) and are strongly metamictic. All but one of the analyzed
grains are slightly to significantly discordant, thus not allowing the calculation of a
concordia age. A
206
Pb/207Pb weighted mean calculation for 16 grains that are less than
27
10% discordant yields an age of 1732±24Ma (MSWD=0.66), which is interpreted as the
age of crystallization of the syenogranite. As in the previous sample, no inheritance was
observed in the grains selected for analysis.
The age of the metamorphic event that affected the granitoids of the Araracuara area is
constrained between the crystallization ages discussed above and K-Ar cooling ages of
around 1320-1340 Ma obtained in micas from nearby rocks (samples PRA-51 and PRA53 of Priem et al. 1982). No precisions about the age of metamorphism can be made
based on the zircon U-Pb data presented here.
5.3.2. Mitu-Taraira Granitoids:
Granitoids selected from the Mitu and Taraira regions of Colombia are all
petrographically very similar, characterized by porphyritic granitoids with subordinate
mafic minerals like biotite and amphibole as the only ferromagnesian-bearing phases.
Perthitic textures are common in the feldspars and amphiboles display optical properties
of the Riebeckite-Arfvedsonite series, indications of the strong alkalic character of these
intrusives. Zircons from all the samples are mostly prismatic displaying width:length
ratios between 1:3 and 1:7, with U contents that typically range from ~60 to ~1500 ppm.
Inheritance is common in all the samples.
Crystallization ages calculated from these granites are also very similar, falling within a
restricted range from 1530 to 1588 Ma (Figures 7c, d, e, f). Sample AH-1216 from the
Vaupes River SE of Mitu yields a concordia age of 1574±16 Ma (MSWD=0.37)
calculated from seven concordant analyses, with most of the inherited cores defining a
rather coherent population at ~1780 Ma. Most of the analyzed spots from the igneous
portions of the grains are affected by modern Pb-loss and are shown in gray in Figure 7c.
A 206Pb/207Pb weighted mean age calculated using all the analyzed spots, including those
that are >5% discordant, produces an undistinguishable age of 1574±17 Ma with an
MSWD=0.20. Sample PR-3092 from the Lerida settlement along the Apaporis River
produced a complex concordia, as strong metamictization of the grains resulted in
significant Pb-loss and discordance. Nine grains that are less than 10% discordant define
28
a
206
Pb/207Pb weighted mean age of 1578±30 Ma (MSWD=0.76), and most of the
inherited grains cluster around ~1800 Ma. This sample has the oldest zircons analyzed in
the present study, with three xenocrysts that yield early Paleoproterozoic ages between
2.1-2.3 Ga and two others with late Archean ages around 2.6 Ga. Zircons from sample
AH-1419, collected from the Raudal Cupay in the Apaporis River proven to be mostly
concordant (Figure 7e). Analyses from 11 igneous crystals and/or overgrowths yields a
206
Pb/207Pb weighted mean age of 1530±24 Ma (MSWD=0.32) while most of the
xenocryst cores have ages between 1.70 and 1.86 Ga. There is another set of inherited
ages that defines a younger cluster at ~1.59 Ga. Finally, sample CJR-19 from the Raudal
La Libertad on the Apaporis River yields a 206Pb/207Pb weighted mean crystallization age
of 1588±21 Ma (MSWD=1.03) calculated using 35 analyses performed in oscillatoryzoned grains and overgrowths as revealed by CL imaging. Analyses performed in areas
interpreted as inherited xenocrysts are in some cases as old as ~1.9Ga, but some others
show a significant age overlap with spots interpreted as reflecting granitoid
crystallization. However, age calculation using the zircon-age extractor algorithm of
Isoplot results in an undistinguishable age of 1591 +20/-4 Ma, obtained from a coherent
group of 44 analyses. It is possible that some of the xenocryst analyses that are slightly
older than the crystallization ages only represent mixed ablation spots between the
xenocryst and the magmatic overgrowths. This sample corresponds to the series of
granites that conform the basement for the metasediments of the Serrania del Taraira,
thus placing a maximum age of 1.58 Ga for the deposition of these supracrustal
sequences. These stratigraphic and age constraints preclude correlations that suggest
affinity of these metasedimentary sequences with the Roraima Group (Galvis et al.,
1979), given its exclusively Paloproterozoic age (Santos et al., 2003).
6. Discussion. Proposed evolution of the Putumayo orogen within a Rodinian
context:
The participation of the Amazon Craton during the assembly of the Rodinia
supercontinent has for long been inferred, using the Bolivian-Brazilian Sunsas-Aguapei
orogenic belt as a tie to other Grenville-age orogenic segments worldwide (Cordani et al.
2009; Hoffman, 1991; Li et al. 2008). Continuation of this Meso- Neoproterozoic belt
29
towards the northwest extending beneath the Amazon River graben and the north Andean
foreland basins has commonly been inferred by some authors, but this assumption has so
far only been based on indirect evidence. Observations used to support this idea are 1) the
widespread occurrence of ~1Ga zircon populations in parautochthonous lower Paleozoic
sequences of the Peruvian Andes (Cardona et al. 2009; Chew et al. 2007, 2008; Cordani
et al. 2010), and 2) the occurrence of Grenville-age inliers in the Andes of Colombia
(Fuck et al. 2008; Li et al. 2008; Restrepo-Pace et al. 1997). However, the complex
tectonic evolution of the northern Andes characterized by semi-continuous subduction
and superposed extensive terrane transfer and accretion throughout the Phanerozoic
(Aspden et al. 1987; Bayona et al. 2006, 2010; Forero-Suarez, 1990; McCourt et al. 1984;
Ramos, 2010; Toussaint, 1993), casts doubts on the direct correlation between the
basement blocks exposed in the cordillera and that of its related foreland basins. The
geochronological results presented in this paper provide, for the first time, direct evidence
that demonstrates the existence of an autochthonous Grenville-age orogenic belt along
the northwestern margin of Amazonia, which serves both as a basis for global
intercratonic correlations and as a reference against which the Cordilleran inliers and
other supposedly peri-Amazonian blocks can be compared. As will be further expanded
below, our interpretations of the geological and geochronological data obtained from this
study favor a tectonic model for the evolution of NW Amazonia’s Mesoproterozoic
margin that is very different from that of the Sunsas-Aguapei orogen. Consequently, we
propose the name Putumayo orogen for a distinct segment of a Grenville-age mobile belt
in northern South America, whose paleogeographic reconstruction and global tectonic
implications should also be different to the Sunsas belt, according to the nature and
timing of the major geologic events that locally characterized its evolution. We propose a
two-stage metamorphic history for the development of this orogen, characterized by
initial accretionary orogenesis followed by final ocean closure and continent-continent
collision. The following discussion will be subdivided in four temporal intervals, roughly
following the paleogeographic model presented in Figure 10. They include: 1) PreRodinian framework: the peri-Amazonian “Oaxaquian-Colombian” arc phase, 2) Rodinia
assembly initiation: early arc and terrane accretions onto Amazonia’s margin, 3) Rodinia
final assembly: the Putumayo-Sveconorwegian orogeny; and 4) some observations on
30
regional cooling and post-collisional evolution. A correlation chart comparing the timing
of major geological events in the different blocks considered for the tectonic
reconstruction and discussed throughout the text is presented in Figure 9.
6.1. Pre-Rodinian framework (~1.3-1.08 Ga):
The position of Amazonia during this time period is constrained by few paleomagnetic
poles obtained in the SW portion of the Amazon craton that suggest an oblique collision
between Laurentia and Amazonia along the Llano-Rondonia segments at ca. 1.2 Ga
(Tohver et al. 2002), and a continued strike-slip displacement that juxtaposes the same
SW margin of Amazonia against the SE Appalachian margin at ca. 1.15 Ga (D'AgrellaFilho et al., 2008). Additional geochemical/geochronological evidence supporting this
model is the presence of exotic, Amazonian-affine crust in the Blue Ridge /Mars hill
inlier of the southern Appalachians (Loewy et al. 2003; Tohver et al. 2004a). Structural
data for both of the supposedly colliding margins agrees with a sinistral-shearing
kinematic history in both the Llano and the Blue Ridge segments, the later coinciding
with the final inversion of the Nova Brasilandia intracontinental rift of central-western
Brazil (Tohver et al. 2004b). Significant deformation in both segments is culminated by
the occurrence of undeformed, postectonic intrusions, represented by the Sunsas granite
suite dated at 1076±18 Ma in the Sunsas-Aguapei belt (U-Pb SHRIMP age by Boger et
al. 2005) and undeformed rhyolitic dikes dated at 1098±10 Ma by ID-TIMS in the eastern
Llano uplift (Mosher, 1998; Walker, 1992).
The relative position and interaction of Baltica with respect to Laurentia during this
period is better understood, constrained by both paleomagnetic and geologic data
(Cawood and Pisarevsky, 2006; Cawood et al. 2010; Pisarevsky et al. 2003 and
references therein). Prior to ~1.25Ga, Baltica is though to have been attached to the
modern eastern coast of Greenland along the northern Timanian margin (Figure 10a),
before it started a late Mesoproterozoic ~90 degrees clockwise rotation on its way to its
final Rodinia position (Cawood and Pisarevsky, 2006). This rifting and rotation is
responsible for the opening of the so-called Asgard sea, where a subduction complex was
established along the Laurentian margin giving birth to the Neoproterozoic Valhalla
31
orogen (Cawood et al. 2010). During this rotation, a passive margin system is established
along Baltica’s trailing edge and accretionary tectonics were triggered along its leading
edge during the early stages of the Sveconorwegian orogeny (Bingen et al. 2008b;
Bogdanova et al. 2008).
6.1.1. Geochronological structure of Amazonia’s continental leading margin:
U-Pb zircon geochronological results presented in this study confirm the early Meso- to
late Paleoproterozoic character of the crystalline basement exposed in the eastern
Colombian lowlands. Undeformed alkaline granitoids with crystallization ages between
1.53-1.58 Ga intrude a basement that is predominantly equal or older than ca. 1.7Ga old,
considering the ages of inherited zircon xenocrysts in the Mesoproterozoic intrusives
(Figure 7) and in accordance with a SHRIMP U-Pb age of 1703±7 Ma from a nearby
biotite qz-diorite along the Colombian-Brazilian border (Tassinari et al. 1996). Further
west, in the Araracuara basement window, deformed granitoids yield U-Pb ages of
1732±24 and 1756±18 Ma, evidencing the extent of the Paleoproterozoic basement in
eastern Colombia. All these results are broadly in agreement with a NW extension of the
Rio Negro-Juruena province from Brazil towards Colombian territory (Tassinari and
Macambira, 1999), where early Mesoproterozoic undeformed granites intrude
Paleoproterozoic gneisses and deformed granitoids. These intrusive suites are possibly
similar in nature to the ~1.5 Ga anorogenic granitic complexes that are widespread in the
RNJ province and western Amazonia in general (e.g. Parguaza Batholith in Venezuela,
Gaudette et al. 1978; Rio Crespo intrusive suite of Rondonia, Bettencourt et al. 2010;
among many others as summarized by Dall'Agnol et al. 1999).
Furthermore, latest Paleoproterozoic protolith ages obtained from drilling-well cores of
migmatitic gneisses underlying the Putumayo foreland (wells Mandur-2 and Payara-1,
Figure 6) suggest that, previous to the inception of the Putumayo orogen, the
Paleoproterozoic basement of NW Amazonia extended all the way towards the cratonic
margin (today’s limits with the Cordillera, covered under the Andean foreland). This
observation contrasts with the regional structure mapped in SW Amazonia, where the
Rondonia-San Ignacio Mesoproterozoic orogen lies outboard of the Rio Negro-Juruena
32
province and separates it from the Sunsas-Aguapei orogen (Cordani and Teixeira, 2007).
It is interesting to note the conspicuous lack of evidence for zircon-forming events
throughout most of the Mesoproterozoic in samples analyzed from the Putumayo
basement. While during this period SW Amazonia’s margin is characterized by
voluminous arc-related magmatism and crustal growth by accretion of island arcs and
continental terranes (Bettencourt et al. 2010; Cordani and Teixeira, 2007), autochthonous
NW Amazonia seems to have experienced a completely different geological history in a
tectonically more stable setting, conceivably as a wide back-arc or a passive margin.
6.1.2. Basins and sedimentation in Amazonia’s continental margin:
The geological knowledge of the Proterozoic basins and sedimentary successions in NW
South America is very limited. The only Proterozoic sedimentary units that have been
described to date are metasediments from the Taraira, Caranacoa and Naquen ranges of
eastern Colombia (of unknown age, but arguably younger than 1.5Ga given the new ages
presented here from granites of the Taraira area), and the volcanosedimentary Piraparana
Formation of Priem et al. (1982). Cawood et al. (2007) suggested a possible correlation
between the Piraparana formation of eastern Colombia with cratonic-cover and rift
sequences of Rondonia (Huanchaca and Aguapei groups). However, these sequences in
the Bolivia-Brazil area are interpreted as intracratonic or supracrustal in nature, deposited
in an intracontinental-rift and/or passive-margin settings prior to the Sunsas collisional
phase (Teixeira et al. 2010). Rhyodacitic lavas included in the Piraparana formation have
been dated at ca. 920 Ma by Rb-Sr (Priem et al. 1982), younger that the depositional ages
inferred for the passive-margin and rift sequences in the Sunsas region and postdating
collisional orogenesis recorded in the Putumayo orogen. Instead of a marginal cratonic
cover, we propose that the Piraparana formation could represent either 1) foreland basin
deposits related to Putumayo orogenic development inboard in Amazonia, or 2)
Neoproterozoic syn-rift sedimentation and volcanism associated with early extensional
events of the Neoproterozoic Güejar-Apaporis graben preceding the collapse of the
Putumayo orogen and related Grenville-age belts. Only detailed sedimentary provenance
studies in the Piraparana formation will allow us to test these hypotheses.
33
Metasedimentary migmatites recovered in the Caiman-3 well show detrital zircon age
spectra indicative of provenance from the cratonic interior, dominated by Rio NegroJuruena and Ventuari-Tapajos -like ages (Figure 8). Distribution peak maxima at ca.
1470, 1500, 1570, 1650-1670 and 1880 Ma, with more than 50% of the analyzed grains
on the Paleoproterozoic side of the spectrum, are completely different from the patterns
observed in metasediments associated with the Cordilleran terranes, where no grains
older than 1.51 Ga were found in the more than 400 individual zircons analyzed.
Maximum depositional ages for the sedimentary protolith of these migmatites are
calculated at 1444±15 Ma (1σ, total error) from the age of the youngest grain of the
>95% concordant xenocryst analyses. Although the age of deposition of the protolith is
loosely bracketed between this age and the metamorphism at 989±14 Ma (Figure 6e),
sedimentation clearly post-dates the protoliths of the Mandur-3 and Payara-1 gneisses,
implying that these migmatites represent a metamorphosed Mesoproterozoic sedimentary
cover of the Paleoproterozoic marginal basement in the area. It is also entirely possible
that deposition of this sedimentary sequence occurred later than 1.44 Ga, in a location
where syn-sedimentary arc influence was negligible or non-existent and only cratoniclike ages are recorded in the zircon age spectra.
6.2. The Oaxaquian-Colombian fringing-arc system (1.30-1.10 Ga):
Proterozoic basement blocks found in Mexico are included in the units known as the
Novillo gneiss, Huiznopala gneiss, the Oaxacan complex, and the Guichicovi complex
(Figure 1 in Keppie and Ortega-Gutierrez, 2010). These inliers are characterized by a
similar Proterozoic history consisting in intrusion of granitoids with arc and backarc
chemical affinities during the interval ~1.3 to 1.15 Ga, intrusion of AMC (AnorthositeMangerite-Charnokite) complexes around 1.01 Ga, and high-grade granulite-facies
metamorphism at around 0.99 Ga (Cameron et al., 2004; Keppie and Ortega-Gutierrez,
2010; Solari et al. 2003; Weber and Hecht, 2003; Weber and Kohler, 1999). In contrast to
Mexico, few Proterozoic meta-igneous units have been identified in the Colombian
Andes, where they are only represented by the Guapoton orthogneiss (Cordani et
al.,2005), Las Minas orthogneiss and La Macarena mylonites (this study), and the San
Lucas orthogneiss of Ordoñez-Carmona et al. (2009). Most of the Proterozoic units
34
outcropping in the Colombian inliers are metasedimentary or metavolcanic in nature
(Kroonenberg, 1982). Similarities in Pb isotopic compositions, U-Pb ages of granulitefacies rocks, and Paleozoic fossil faunas from Colombia and Mexico, led Ruiz et al.
(1999) to suggests a correlation between the Proterozoic inliers of these two regions
specially between the Garzon, Santa Marta and Guichicovi inliers, correlation that has
found further support with recent zircon Hf isotopic data (Weber et al. in press).
In many Mesoproterozoic paleogeographic reconstructions, Oaxaquia is commonly
positioned along the northwestern margin of Amazonia in close association with the
Colombian Andean inliers, where it is thought to have evolved as a parautochthonous arc
complex latter to be trapped between colliding continental blocks during the assembly of
Rodinia (Bogdanova et al. 2008; Cardona et al. 2010; Keppie et al. 2001; Keppie and
Dostal, 2007; Keppie and Ortega-Gutierrez, 2010; Keppie and Ramos, 1999; Li et al.
2008). Figure 11 presents a schematic reconstruction for the tectonic configuration of
proto- NW South America, in which we include the Colombian cordilleran inliers and the
Mexican terranes as part of a commonly evolving “Oaxaquian-Colombian” periAmazonian fringing arc system located outboard of Amazonia’s leading edge (Figure
10a). According to Busby et al. (1998), long-lived convergent margins facing a large
ocean basin may evolve through three distinct phases, (1) a strongly extensional
intraoceanic-arc phase, (2) a mildly extensional arc system, and (3) a compressional
continental arc phase following accretion of the fringing-arc terranes. We propose that
the Oaxaquian-Colombian fringing arc may have evolved in a similar fashion, as many of
these stages can be identified from its Mesoproterozoic geological record. Initiation of
the strongly extensional arc phase is represented by the occurrence of 1.3 Ga rift related
basalts (now transformed into mafic granulites) of the north Oaxaca complex (Keppie and
Dostal, 2007) and bimodal igneous suites with backarc geochemical signatures of the
Novillo gneiss (Keppie and Ortega-Gutierrez, 2010). Rifting from Amazonia was shortly
followed by the development of an intraoceanic arc (phase 2), characterized by
widespread arc-related juvenile magmatism between 1.3 and 1.2 Ga in Oaxaquia (Keppie
et al. 2001; Keppie and Ortega-Gutierrez, 2010; Weber and Hecht, 2003; Weber and
Kohler, 1999)Weber et al. in press), which then evolved towards a more mature arc as
35
reflected by granites displaying more crustal isotopic compositions like the 1.15 Ga
protolith of the Guapoton orthogneiss in the Garzon massif (Restrepo-Pace et al. 1997).
Associated to this phase may start the accumulation of thick vulcanosedimentary
sequences included within the Garzon group, now represented by thrust slices of the El
Vergel migmatites in which metasedimentary felsic granulites yield calculated maximum
depositional ages around 1.12 Ga (e.g. sample MIVS-11). Metasedimentary units
included in the Garzon and Las Minas massifs of southern Colombia all have maximum
protolith depositional ages within the range 1.0 to 1.12 Ga, and they are interpreted as
syn-collisional to slightly pre-collisional basins (similar to Assemblage 1 and 2 basins of
Cawood et al. 2007) developed in close proximity to a convergent margin. Detrital zircon
signatures from these units can be explained in terms of derivation from granitic plutons
included within the same Oaxaquian-Colombian arc terrane, with no detritus sourced
from continental Amazonia crust evidenced by the paucity of older than 1.5 Ga (Rio
Negro-Juruena –like) ages observed in the spectrum (Figure 8). The absence of
Paleoproterozoic detritus in these units, both pre- and syn-collisional could be explained
in at least two ways: First, in terms of a high-standing continental arc acting as a
topographic barrier preventing detritus of the cratonic interior from reaching a forearc
basin, and 2), that there was truly a disconnect in the sedimentary systems between arcrelated basins and those formed along the stable continental margin. In the second case,
these basins could conceivably be separated by a wide back-arc developed behind a
fringing-arc complex (Figure 11). It should be noted that there is a slight overlap in the
age distribution spectra between the metasediments with craton-like signatures of the
Caiman-3 migmatites discussed above and those from the Garzon and Las Minas massifs,
spanning the age range from 1.46 to 1.50 Ga (Figure 8). There are many lines of evidence
that support the presence of older continental material to be present within the
parautochthonous terranes; Recently reported orthogneisic units with protolith ages
around 1.5 Ga in the San Lucas Massif (Ordoñez-Carmona et al. 2009), and protolith
crystallization ages of 1.46 Ga from metaigneous mylonites of the Macarena range from
this study, comprise direct evidence for the presence of such old blocks within the
Colombian terranes. We suggest that initial rifting of the Oaxaquian-Colombian arc from
Amazonia during the initial extensional phase may have detached fragments of older
36
continental crust and incorporated them within the basement of the intraoceanic arc
(Figure 11), so differences in Nd and Hf isotope compositions between the different
orthogneissic units could also be explained in terms of different crustal contamination
histories related to along-strike variations in the nature of the basement that makes up
each segment of the arc (i.e. segments built up only by juvenile rocks vs. segments which
have incorporated older Proterozoic basement of Amazonia following initial arc rifting).
Based on Hf isotope systematics in zircons, Weber et al. (submitted) identified two
different reservoirs that took part in the evolution of Oaxaquia: First, what they term
“typical Oaxaquia”, comprised by juvenile intrusives of an early intraoceanic arc which
yield high positive εHf initial values (orthogneisses of the Oaxaca complex and the
Novillo gneiss), and second, crustal slices with an older cratonic Amazonian component
that yield mixed initial εHf compositions between a juvenile and a more evolved
component (the Huiznopala and Guapoton orthogneisses). Following from this
distinction, these authors propose that typical Oaxaquia crust evolved as an intraoceanic
arc outboard of Amazonia, whereas the less radiogenic arc-related intrusives of the
Huiznopala and Guapoton orthogneisses were developed along and active continental
margin built on cratonic Amazonia crust. Geochronological results from core samples
recovered in the Putumayo basin basement presented in this study do not support the
existence of a late Mesoproterozoic active margin developed in autochthonous Amazonia
crust. Instead, our data suggest that the cratonic margin of northwestern South America is
characterized by late Meso- to early Neoproterozoic metamorphic reworking of late
Paleoproterozoic basement and associated Mesoproterozoic cover sequences, with no
magmatic events identified within this range. In our reconstruction (Figure 10), we
include all the late Mesoproterozoic Colombian and Mexican inliers in a common
fringing arc system outboard of Amazonia instead of separated in two sub parallel arcs.
Following this period of intensive arc plutonism, there is an interval of magmatic
quiescence that spanned the age range between 1100 and 1050 Ma and is present both in
Oaxaquia and the Colombian terranes (Figure 9). One possible scenario for the sudden
shutting down of magmatism could be a drastic change of stress regimes along the arc,
37
switching from a mildly extensive to a compressive regime (moving from phase 2 to
phase 3) that could have resulted in increased plate coupling along the margin, continentdirected trench migration, and final collapse of the fringing arc against the continental
margin (e.g. Busby, 2004). This transition between intraoceanic arc magmatic activity
and terrane accretion marks the initiation of collisional events recorded in the Putumayo
orogen.
6.3. Rodinia assembly initiation: terrane and arc accretion onto Amazonia’s margin
(1.07-1.01 Ga):
Although mostly obliterated by the later ca. 0.99 Ga granulite-facies overprint, there is
textural and geochronological evidence supporting the existence of an early metamorphic
event between 1.05 and 1.01 Ga, both in the cordilleran inliers (Cordani et al. 2005;
Jimenez Mejia et al. 2006) and in the basement of the Putumayo basin (this study).
Migmatization of the Las Margaritas unit in the eastern side of the Garzon massif is
constrained to 1015±7.8 Ma by SHRIMP U-Pb dating of zircons from a leucosome and
1034±16 Ma by a Sm-Nd garnet-whole rock isochron, although additional Sm-Nd data
evidences that the ~990 Ma event is locally present within this unit as well (Cordani et al.
2005). In the basement of the Putumayo basin, the amphibolite-facies metamorphic event
identified in the Mandur-2 and Solita-1 wells yields ages of 1017±10 Ma and 1044±26
Ma, respectively (Figure 9), and in the Guamoco area of the Colombian Central
Cordillera just south of the Serrania de San Lucas an orthogneiss unit with protolith ages
of ca. 1.28 Ga was also affected by metamorphism at 1.04 Ga (Victor Valencia, personal
communication). Late Mesoproterozoic tectonothermal events constrained between 1.1
and 1.07 Ga are also known from the northern Oaxaca complex (Solari et al. 2003);
Weber et al. submitted), where they have received the name of Olmecan orogeny. These
migmatization ages are interpreted in terms of accretion of the Oaxaquian-Colombian arc
onto Amazonia continental margin, probably along a series of slightly diachronous softcollision events. Amalgamation of the fringing terranes against Amazonia was possibly
shortly followed by slab-break off and melting of the lower crust, leading to the
generation of AMC intrusive complexes that are widespread amongst the Oaxaquian
38
terranes (Weber et al. submitted) and which are possibly also recorded in the Sierra
Nevada de Santa Marta massif in the northernmost Colombian Andes.
In Baltica, metamorphism of the Bamble and Kongsberg tectonic wedges marks the
timing of early terrane accretions onto Fennoscandian crust through the collision of
Telemarkia against the Idefjorden terrane (the Arendal phase, Bingen et al., 2008b). The
Bamble and the Kongsberg terranes comprise variably metamorphosed supra-subduction
complexes whose metamorphic ages constrain the age of collision between ~1140 and
1090 Ma, earlier than the initial accretions along Amazonia’s margin as recorded in the
Putumayo Orogen and Oaxaquia. Additionally, marked differences that exist between the
detrital zircon age spectra of Telemarkia’s metasedimentary units (Bingen et al. 2001;
Bingen et al. 2003), and those from NW autochtonous/parautochtonous Amazonia
(Cardona et al. 2010; this study) lead us to conclude that regardless of the exotic or
indigenous origin of Telemarkia with respect to Fennoscandia, an Amazonian ancestry
for this terrane and its subsequent accretion to Baltica by early collisional interactions
with Amazonia at ca. 1.1 Ga are both unlikely.
6.4. Final Rodinia assembly: maximum convergence along the PutumayoSveconorwegian orogen (1.00-0.98 Ga):
As it was mentioned before, collisional interaction between Baltica and Amazonia during
the early Neoproterozoic agglomeration of Rodinia have already been suggested by
previous authors (Bingen et al. 2008b; Bogdanova et al. 2008; Cardona et al. 2010;
Keppie and Dostal, 2007; Li et al. 2008). However, geological evidence to support this
correlation was previously not available from the Amazonian side, given that the only
previously known outcrops of Grenville-age crust were Cordilleran inliers whose
autochthonous character with respect to non-remobilized NW Amazonia was uncertain.
New U-Pb data presented here demonstrates the existence of a Grenville-age orogenic
belt in NW South America, buried under Andean foreland basin sequences, and for the
first time provides a link between granulite-facies tectonothermal events occurring in NW
cratonic Amazonia, the Colombian inliers, Oaxaquia, and the Sveconorwegian orogen
(Figure 9).
39
As noted by Bogdanova et al. (2008), paleogeographic reconstructions between Laurentia
and Baltica imply the necessity of a “third continent” in order to explain some periods of
the Sveconorwegian orogeny, and they suggest collision with Amazonia as a feasible
tectonic scenario. Additionally, these authors also point that the chronology of the final
granulite-facies event in the Sveconorwegian indicates that crustal shortening was still
underway along this orogen while extensional regimes were already dominating the
Grenville margin (Gower and Krogh, 2002; Gower et al. 2008), so in addition of the third
continental mass needed in order satisfy late-Sveconorwegian orogenic events, this third
continent needs to fulfill certain geological requirements in terms of the chronology for
the final collision-related metamorphism. All of the Oaxaquian inliers show evidence of
an early Neoproterozoic granulite-facies tectonothermal event at ca. 990 Ma (Cameron et
al. 2004; Keppie and Ortega-Gutierrez, 2010; Lawlor et al. 1999; Ruiz et al. 1999; Solari
et al. 2003; Weber and Kohler, 1999) known as the Zapotecan event, which is also
coincident with the age of the granulite-facies metamorphism from the Garzon Massif (El
Vergel granulites, Figure 3a and c; Guapoton-Mancagua orthogneiss, Figure 3b), Las
Minas orthogneiss (Figure 3d), and is also recorded in ortho- and para-migmatites of the
Putumayo basin basement (Payara-1 and Caiman-3 wells, Figure 6). In line with these
chronological evidences, our reconstruction displays final collisional between Baltica and
Amazonia along a juxtaposed Sveconorwegian-Putumayo orogenic belt at ca. 990 Ma,
defining a separate belt with a structural trend that lies almost orthogonal with respect to
the Sunsas-Grenville orogenic tract (Figure 10). Continuation of the Putumayo Orogen
towards the (actual) northeast, underlying the foreland Llanos basins of northern
Colombia and Venezuela still awaits for conclusive evidence, although there are
observations that may suggest it is so. Occurrences of possible Grenville-age orogenic
rocks in Venezuela are expressed as: 1) The presence of the high-grade “Nirgua
complex” inlier of the southern Merida Andes, consisting of mafic and felsic
charnockites crosscut by ultramafic dykes resembling granulitic complexes of the
Colombian Andes, and 2) A wide variety of high-grade rocks that occur as xenoliths in
Oligo-Miocene diatremes, in drilling wells from the La Vela Gulf, and boulders in
Paleocene and Miocene conglomerates (Grande and Urbani 2009 and references therein).
40
Additionally, the occurrence of feldspar K-Ar cooling ages of ca. 1.0 Ga in samples from
drilling-well cores in eastern Venezuela (locations in Figure 1) also records the extension
of the Grenville-age heat pulse towards this portion of South America (Feo-Codecido et
al. 1984)
Cameron et al. (2004) noted that there is no obvious variation in ages for the granulitefacies event in Oaxaquia with respect to geographical position, observation that also
holds for the available U-Pb ages of high-grade metamorphism in the Colombian massifs
and the Putumayo basin basement. With the exception of the Zancudo migmatites (CB006, Figure 3e), ages presented in this study for the early-Neoproterozoic metamorphic
event all fall in a restricted range between 986 and 992 Ma, indistinguishable from each
other within analytical error. These observations suggest that the character of the
collision between Baltica and Amazonia was relatively close to frontal between the protomargins. Following collision, shortening and crustal thickening may have continued in
the orogen, identified in the Sveconorwegian belt as continued deformation along the
mylonite zone between 980-970 Ma, propagation of the foreland, and eclogite-facies
metamorphism on the “eastern segment” at 970 Ma (Johansson et al. 2001), also
coinciding with the early stages of intrusion of mafic dikes related to the BlekingeDalarna dolerite swarm (Soderlund et al. 2005). Late tectonothermal events with similar
ages are also recorded along the margin of Amazonia, where metatexic migmatites of the
Las Minas massif yield ages of 972±14 for the anatectic event, similar to other ages of
~0.97 found elsewhere in Oaxaquia (Cameron et al., 2004; Weber et al. in press). In the
Putumayo basin, the undeformed leucogranite that crosscuts the foliation of the
migmatites in the Caiman-3 well provides a minimum age of 952±21 Ma for pervasive
deformation in this sector of the Putumayo orogen. This age is similar to that obtained for
undeformed granitic pegmatites in Oaxaquia, constrained to ca. 0.97 Ga as discussed by
Solari et al. (2003).
6.5. Post-collisional events and regional cooling (post 0.97 Ga):
In different locations of the Colombian Andes, medium- to low-temperature cooling of
the high-grade basement of the cordilleran inliers is constrained by K-Ar and Ar-Ar data.
41
Although ages spread over a wide range from ~1028 to ~561 Ma, most of them cluster in
a restricted range from 970 to 910 Ma (Cordani et al. 2005; Restrepo-Pace et al. 1997)
implying relatively fast rates of exhumation for the orogenic rocks. Metamorphic rims in
zircons from the Santander and Jojoncito paragneisses yield U-Pb ages of metamorphism
at 864±66Ma and 916±19Ma respectively (Cordani et al. 2005), much younger than any
of the other Proterozoic metamorphic events recorded elsewhere in the Colombian
Andes. Although inherited zircon analyses from the detrital protolith of these gneisses
display age peaks that resemble those from other metasedimentary units presented in this
study there is an important difference, and its the presence of detrital grains with ages
equivalent to the metamorphic events described from the Garzon and Las Minas massifs
at ca. 980 and 1020-1050 Ma. We suggest that the sedimentary protoliths of these
paragneisses were deposited later than ~970 Ma in basins internal to the orogen
(equivalent to assemblage 3 basins of Cawood et al. 2007), by reworking of older
Putumayo metaigneous and metasedimentary units such as those found in the Garzon and
Minas massifs. In that case, low-P/high-T metamorphic events at ca. 920 and 860 Ma
could be interpreted as related to early phases of crustal extension within the mountain
chain, possibly in a basin and range-like setting during initial orogenic collapse. An
analog for this may be the case of the Swedish “eastern segment” where steep normal
faults and core-complex structures developed between 960 and 900 Ma recording pulses
of extension and collapse in the Sveconorwegian orogen (Bingen et al., 2008b).
6.6. Implications for the peri-Gondwanan realm:
The existence of a Grenville-age Putumayo Orogen in NW South America has
implications for the identification and tracing of peri-Gondwanan blocks in different
ways that will only be briefly mentioned here. 1) It provides a cratonic reference for
evaluating the affinity of crustal blocks with inferred Amazonian ancestries. For example,
it strengthens reconstructions that correlate the Proterozoic inliers of Mexico and
Colombia generating a direct link to autochthonous Amazonia crust. 2) It provides a
feasible source area for the origin of Grenville-age detritus found in blocks that formed
part of or may have interacted with northern Amazonia at some point in their history (e.g.
Marañon complex, Cardona et al. 2009; Maya block, Weber et al. 2008; Ganderian-type
42
massifs, Nance et al. 2007, Finnmarkian Nappes, Corfu et al. 2007, among others). For
example, in an analog way as the Grenville-Sveconorwegian belt shed sediments to the
developing Valhalla orogen (Cawood et al. 2010), the Putumayo orogen and its possible
northwestward extension towards Venezuela could elegantly explain the spectrum of
Grenville-age zircons found in Ganderian-type massifs like NW Iberia / Cadomia, and
could also represent a proximal source for early Neoproterozoic detrital white-micas
found in these massifs (Gutierrez-Alonso et al. 2005; Nance et al. 2007). Given the
differences in timing of the geological events that affected Amazonia during the MesoNeoproterozoic, further refinement of the geochronological database could potentially
allow identifying distinctive fingerprints to separate between Sunsas-like vs. Putumayolike signatures for Amazonia’s Grenville-age detrital zircon fingerprint.
7. CONCLUSIONS
Zircon U-Pb geochronological data presented in this contribution comprises the first
evidence for the existence of a Meso- Neoproterozoic (Grenville-age) orogenic belt in
autochthonous NW Amazonia, buried underneath Meso-Cenozoic sedimentary sequences
of the north Andean Putumayo foreland basin. Samples distributed amongst the Garzon
and Minas massifs and the Putumayo foreland basin basement enclose an area of at least
~50.000 Km2 that are underlain by Grenville-age crust, expanding the known extent of
metamorphic and deformational effects in northern South America caused by continental
collisions during Rodinia assembly. Although many details about the geochemistry,
metamorphic petrology and medium- to low-temperature history of the proposed
Putumayo Orogen are still to be discovered, the new geochronological data obtained
allowed us to identify two distinct tectonometamorphic events of arc-continent and
continent-continent collision comprising its evolution. In addition to the differences in
timing for the peak metamorphic event with respect to the Sunsas-Aguapei orogen, the
conspicuous lack of evidence for Mesoproterozoic zircon-forming events in the
Putumayo
basement,
in
turn
characterized
by
Grenville-age
reworking
of
Paleoproterozoic crust, sets clear differences with respect to the Mesoproterozoic
evolution of the SW Amazonian margin. As implied in the paleogeographic
reconstruction presented in Figure 10, it is suggested that the Guayaquil deflection of the
43
Andean orogen may in some way represent a feature inherited from the Proterozoic
tectonic configuration of Amazonia (an hypothetical Guayaquil promontory?), where it
may have corresponded to a Mesoproterozoic oroclinal bend that separated the Sunsas
margin colliding with Laurentia and the Putumayo margin colliding against Baltica. It
should be noted however, that data presented in this study does not necessarily preclude
correlations that suggest a long-lasting connection between Amazonia and Baltica
throughout the Mesoproterozoic and part of the Paleoproterozoic (i.e. the SAMBA
connection. Johansson, 2009), but we chose a paleogeographic model which is in line
with independent paleomagnetic (D'Agrella-Filho et al., 2008; Tohver et al., 2002),
thermochronologic (Tohver et al., 2006) and isotopic data (Loewy et al., 2003; Tohver et
al., 2004a) that constrain collisional interactions between Laurentia and Amazonia during
Mesoproterozoic times. Our interpretation of the data provides a testable model for the
Proterozoic tectonometamorphic evolution of NW South America, model that will
necessarily face changes and refinements as more detailed petrologic and geothermochronologic information is gathered from this region.
ACKNOWLEDGEMENTS
This research benefited from funding provided by Ecopetrol-ICP, the GSA student grant
#9184-09 and a Chevron-Texaco summer grant through the University Of Arizona. The
authors would like to thank the Colombian “Instituto Nacional del Petroleo-ICP” and the
“Litoteca Nacional de Colombia” for allowing access to the core samples. M.I would
specially like to thank Monica Carvalho for her help with text edition and figure
elaboration, Camilo Bustamante for assistance during fieldwork, and Ana Milena Rangel
for her assistance during core sampling at the Litoteca Nacional. The Laserchron center is
partially funded by NSF-EAR grants #0443387 and #0732436. This paper is specially
dedicated to the memory of Prof. Dr. Thomas Van der Hammen for his inspiring work
about the geology of Colombia and the Amazonas region.
44
FIGURE CAPTIONS
Figure 1. Tectonic map of South America illustrating its major Precambrian constituent
elements with emphasis on the provinces of the Amazon Carton. Modified from Cordani
et al. (2000) and Fuck et al. (2008). The location of the north Andean foreland basins is
highlighted along with the location of drilling-wells from which core samples and
geochronological data have been obtained. Red dot: Location of the core samples
presented in this study. Black dots: core samples presented by other authors; 1, Kovach et
al. 1976; 2, Gorayeb et a. 2004; 3 and 4, Feo-Coecido et al. .
Figure 2. Simplified geological map of our study area in south-central Colombia,
showing the location of samples described in this study. a) Major pre-Mesozoic features
of Colombian geology and location of the Putumayo and Llanos north Andean foreland
basins. Modified after Gomez et al. MAP, ANH b) detailed map of the west Garzon and
Serrania de las Minas area in the Upper Magdalena Valley. Modified from Gomez et al.
Jimenez-Mejia
Figure 3. U-Pb concordia diagrams for samples analyzed from the Garzon and Las Minas
cordilleran inliers.
Figure 4. U-Pb concordia diagram for the sample analyzed from the basement of the
Serrania de la Macarena
Figure 5. Photographs from some of the drilling-core samples analyzed for the present
study. a) Migmatitic gneisses from the Payara-1 well, b) Stromatic migmatites injected by
granitic sills from the Mandur-2 well, c) Metasedimentary felsic migmatites from the
Caiman-3 well, d) Leucogranites that crosscut the metamorphic foliation of the
migmatites in the Caiman-3 well.
Figure 6. U-Pb concordia diagrams for samples analyzed from drilling-core samples of
the Putumayo basin basement.
45
Figure 7. U-Pb concordia diagrams for samples analyzed from outcrops of cratonic
Amazonia in the Araracuara, Mitu and Taraira areas.
Figure 8. Probability plots for the detrital age components of Mesoproterozoic
metasedimentary units included within the Putumayo Orogen, indicating their respective
ages of metamorphism measured by zircon U-Pb and calculated protolith maximum
depositional ages. Shaded red areas represent the spectra plotted from the entire dataset of
each sample while black solid lines show data filtered to <5% discordance. In addition to
the metasedimentary samples, the inherited component of anatectic leucogranites found
in the Caiman-3 is also plotted along with a synthetic diagram that combines all the
individual spot ages analyzed from granitoids of the Mitu-Taraira-Araracuara areas. In
the upper portion of the plot, ranges of relevant magmatic activity from the Proterozoic of
Mexico and Colombia are shown for comparison (data from Mexico Weber , Weber,
Cameron et al. 2004). Colored columns have the same coding as Figure 1 and represent
the age ranges for the evolution of the different provinces of the Amazon craton from
Tassinari and Macambira 1999.
Figure 9. Events correlation chart for different provinces of Amazonia, Baltica and
Laurentia during the time interval from 1200 to 900 Ma, relevant for paleogeographic
correlations discussed throughout the paper. References used are listed in the figure.
(Cardona et al., 2010; Cordani et al., 2005; Ordonez-Carmona et al., 2006; Restrepo-Pace
et al., 1997), (Boger et al., 2005; Cordani et al., 2009; Santos et al., 2008; Teixeira et al.,
2010; Tohver et al., 2004b), (Cameron et al., 2004; Keppie and Ortega-Gutierrez, 2010;
Solari et al., 2003), (Bingen et al., 2008a; Bingen et al., 2008b; Bogdanova et al., 2008),
(Barker and Reed, 2010; Mosher et al., 2008; Reese and Mosher, 2004), (Bartholomew
and Hatcher, 2010; Rivers, 1997)
Figure 10. Geological evolution of the Putumayo orogen from the late Mesoproterozoic
to early Neoproterozoic and proposed paleogeographic correlations between Amazonia,
Baltica and Laurentia. Other blocks and cratons were omitted for implicity. Outline and
46
provinces of Laurentia modified from Cawood et al. (2007); Baltica from Bingen et al.
(2008) and Bogdanova et al. (2008); Amazonia from Cordani et al. (2001). See text for
discussion and other references used.
Figure 11. Geological sketch for the proposed configuration of the OaxaquianColombian fringing-arc system at 1.15 Ga. Modified from Busby et al. (2004)
REFERENCES
Alvarez, J. and Cordani, U.G., 1980. Precambrian basement within the septentrional
Andes: Age and geological evolution, 26ieme Congr. Geol. Int., Paris, pp. 10.
Aspden, J., McCourt, W. and Brook, M., 1987. Geometrical control of subduction-related
magmatism - The Mesozoic and Cenozoic plutonic history of western Colombia. J
Geol Soc London, 144: 893-905.
Barker, D.S. and Reed, R.M., 2010. Proterozoic granites of the Llano Uplift, Texas: A
collision-related suite containing rapakivi and topaz granites. Geol Soc Am Bull,
122(1-2): 253-264.
Bartholomew, M.J. and Hatcher, R.D., 2010. The Grenville orogenic cycle of southern
Laurentia: Unraveling sutures, rifts, and shear zones as potential piercing points
for Amazonia. J S Am Earth Sci, 29(1): 4-20.
Bayona, G. et al., 2010. Paleomagnetic data and K-Ar ages from Mesozoic units of the
Santa Marta Massif: A preliminary interpretation for block rotation and
translations. Journal of South American Earth Sciences, in press.
Bayona, G., Rapalini, V. and Constazo-Alvarez, V., 2006. Paleomagnetism in Mesozoic
rocks of the northern Andes and its implications in Mesozoic tectonics of
northwestern South America. Earth, Planets and Space, 58: 1255–1272.
Bettencourt, J.S. et al., 2010. The Rondonian-San Ignacio Province in the SW Amazonian
Craton: An overview. J S Am Earth Sci, 29(1): 28-46.
Bingen, B., Andersson, J., Soderlund, U. and Moller, C., 2008a. The Mesoproterozoic in
the Nordic countries. Episodes, 31(1): 29-34.
Bingen, B., Birkeland, A., Nordgulen, O. and Sigmond, E., 2001. Correlation of
supracrustal sequences and origin of terranes in the Sveconorwegian orogen of
SW Scandinavia: SIMS data on zircon in clastic metasediments. Precambrian Res,
108(3-4): 293-318.
47
Bingen, B. et al., 2003. Relations between 1.19-1.13 Ga continental magmatism,
sedimentation and metamorphism, Sveconorwegian province, S Norway.
Precambrian Res, 124(2-4): 215-241.
Bingen, B., Nordgulen, O. and Viola, G., 2008b. A four-phase model for the
Sveconorwegian orogeny, SW Scandinavia. Norw J Geol, 88(1): 43-72.
Bogdanova, S.V. et al., 2008. The East European Craton (Baltica) before and during the
assembly of Rodinia. Precambrian Res, 160(1-2): 23-45.
Boger, S., Raetz, A., Giles, D., Etchart, E. and Fanning, C., 2005. U-Pb age data from the
Sunsas region of Eastern Bolivia, evidence for the allochthonous origin of the
Paragua Block. Precambrian Res, 139(3-4): 121-146.
Busby, C., 2004. Continental growth at convergent margins facing large ocean basins: a
case study from Mesozoic convergent-margin basins of Baja California, Mexico.
Tectonophysics, 392(1-4): 241-277.
Busby, C., Smith, D., Morris, W. and Fackler-Adams, B., 1998. Evolutionary model for
convergent margins facing large ocean basins: Mesozoic Baja California, Mexico.
Geology, 26(3): 227-230.
Cáceres, C., Cediel, F. and Etayo, F., 2003. Mapas de distribución de facies sedimentarias
y armazón tectónico de Colombia a través del Proterozoico y del Fanerozoico.
Ingeominas, Bogota, 40 pp.
Cameron, K.L. et al., 2004. U-Pb geochronology and Pb isotope compositions of leached
feldspars: constraints on the origin and evolution of Grenvillian rocks from
eastern and southern Mexico. In: R.P. Tollo, L. Corriveau, J.B. McLelland and G.
Bartholemew (Editors), Proterozoic Tectonic Evolution of the Grenville Orogen
in North America. Geological Society of America Memoir, pp. 755–769.
Cardona, A. et al., 2010. Grenvillian remnants in the Northern Andes: Rodinian and
Phanerozoic paleogeographic perspectives. J S Am Earth Sci, 29(1): 92-104.
Cardona, A. et al., 2009. U-Pb Zircon Geochronology and Nd Isotopic Signatures of the
Pre-Mesozoic Metamorphic Basement of the Eastern Peruvian Andes: Growth
and Provenance of a Late Neoproterozoic to Carboniferous Accretionary Orogen
on the Northwest Margin of Gondwana. J Geol, 117(3): 285-305.
Cardona-Molina, A., Cordani, U.G. and MacDonald, W.D., 2006. Tectonic correlations
of pre-Mesozoic crust from the northern termination of the Colombian Andes,
Caribbean region. J S Am Earth Sci, 21(4): 337-354.
48
Cawood, P.A., Nemchin, A.A., Strachan, R., Prave, T. and Krabbendam, M., 2007.
Sedimentary basin and detrital zircon record along East Laurentia and Baltica
during assembly and breakup of Rodinia. J Geol Soc London, 164: 257-275.
Cawood, P.A. and Pisarevsky, S.A., 2006. Was Baltica right-way-up or upside-down in
the Neoproterozoic? J Geol Soc London, 163: 753-759.
Cawood, P.A. et al., 2010. Neoproterozoic orogeny along the margin of Rodinia: Valhalla
orogen, North Atlantic. Geology, 38(2): 99-102.
Chew, D., Kirkland, C., Schaltegger, U. and Goodhue, R., 2007. Neoproterozoic
glaciation in the Proto-Andes: Tectonic implications and global correlation.
Geology, 35(12): 1095-1098.
Chew, D.M. et al., 2008. Detrital zircon fingerprint of the Proto-Andes: Evidence for a
Neoproterozoic active margin? Precambrian Res, 167(1-2): 186-200.
Collins, A.S., Santosh, M., Braun, I. and Clark, C., 2007. Age and sedimentary
provenance of the Southern Granulites, South India: U-Th-Pb SHRIMP secondary
ion mass spectrometry. Precambrian Research, 155(1-2): 125-138.
Connelly, J., 2001. Degree of preservation of igneous zonation in zircon as a signpost for
concordancy in U/Pb geochronology. Chem Geol, 172(1-2): 25-39.
Cordani, U.G., Cardona, A., Jimenez Mejia, D.M., Liu, D. and Nutman, A., 2005.
Geochronology of Proterozoic basement inliers in the Colombian Andes: tectonic
history of remnants of a fragmented Grenville belt Geological Society, London,
Special Publications, 246: 329-346.
Cordani, U.G., Fraga, L.M., Reis, N., Tassinari, C.C.G. and Brito-Neves, B.B., 2010. On
the origin and tectonic significance of the intra-plate events of Grenvillian-type
age in South America: A discussion. J S Am Earth Sci, 29(1): 143-159.
Cordani, U.G. and Teixeira, W., 2007. Proterozoic accretionary belts in the Amazonian
Craton The Geological Society of America, Special Publications, 200: 297-320.
Cordani, U.G., Teixeira, W., D'agrella-Filho, M.S. and Trindade, R.I., 2009. The position
of the Amazonian Craton in supercontinents. Gondwana Research, 15(3-4): 396407.
Corfu, F., Hanchar, J.M., Hoskin, P.W.O. and Kinny, P., 2003. Atlas of zircon textures.
In: J.M. Hanchar and P.W.O. Hoskin (Editors), Zircon. Reviews in Mineralogy
and Geochemistry. Mineralogical Society of America & Geochemical Society.
Corfu, F., Roberts, R.J., Torsvik, T.H., Ashwal, L.D. and Ramsay, D.M., 2007. PeriGondwanan elements in the Caledonian Nappes of Finnmark, Northern Norway:
49
Implications for the paleogeographic
Caledonides. Am J Sci, 307(2): 434-458.
framework
of
the
Scandinavian
D'Agrella-Filho, M.S. et al., 2008. Direct dating of paleomagnetic results from
Precambrian sediments in the Amazon craton: Evidence for Grenvillian
emplacement of exotic crust in SE Appalachians of North America. Earth Planet
Sc Lett, 267(1-2): 188-199.
Dall'Agnol, R., Costi, H., Leite, A., Magalhaes, M.d. and Teixeira, N., 1999. Rapakivi
granites from Brazil and adjacent areas. Precambrian Research, 95(1-2): 9-39.
Dalziel, I., 1991. Pacific margins of Laurentia and east Antarctica Australia as a
conjugate rift pair - Evidence and implications for an Eocambrian supercontinent.
Geology, 19(6): 598-601.
Feo-Codoecido, G., Smith, F., Aboud, N. and Di Giacomo, E., 1984. Basement Paleozoic
rocks of the Venezuelan Llanos Basin, 162. Geological Society of America.
Forero-Suarez, A., 1990. The basement of the Eastern Cordillera, Colombia - An
allochthonous terrane in northwestern South America. Journal of South American
Earth Sciences, 3(2-3): 141-151.
Fuck, R.A., Brito Neves, B.B. and Schobbenhaus, C., 2008. Rodinia descendants in
South America. Precambrian Res, 160(1-2): 108-126.
Galvis, J., Huguett, A. and Ruge, P., 1979. Geologıa de la Amazonia Colombiana.
Boletin Geologico del Ingeominas, 22(3): 3-86.
Gaudette, H., Mendoza, V., Hurley, P. and Fairbairn, H., 1978. Geology and age of
Parguaza rapakivi granite, Venezuela. Geol Soc Am Bull, 89(9): 1335-1340.
Gehrels, G., Valencia, V. and Pullen, A., 2006. Detrital zircon geochronology by LaserAblation Multicollector ICPMS at the Arizona LaserChron Center. In: T.D.
Olszewski (Editor), Geochronology Emerging opportunities. Paleontological
Society, pp. 67-76.
Gehrels, G.E., Valencia, V.A. and Ruiz, J., 2008. Enhanced precision, accuracy,
efficiency, and spatial resolution of U-Pb ages by laser ablation-multicollectorinductively coupled plasma-mass spectrometry. Geochem Geophy Geosy, 9:
Q03017.
Gómez, J. et al., 2007. Mapa Geológico de Colombia. Escala 1:2'800.000. . Ingeominas,
Bogota.
Gower, C. and Krogh, T., 2002. A U-Pb geochronological review of the Proterozoic
history of the eastern Grenville Province. Can J Earth Sci, 39(5): 795-829.
50
Gower, C.F., Kamo, S. and Krogh, T.E., 2008. Indentor tectonism in the eastern
Grenville Province. Precambrian Res, 167(1-2): 201-212.
Grande, S. and Urbani, F., 2009. Presence of high-grade rocks in NW Venezuela of
possible Grenvillian affinity. In: K.H. JAMES, M.A. LORENTE and J.L.
PINDELL (Editors), The Origin and Evolution of the Caribbean Plate. Geological
Society of London Special Publications, pp. 533–548.
Gutierrez-Alonso, G., Fernandez-Suarez, J., Collins, A., Abad, I. and Nieto, F., 2005.
Amazonian mesoproterozoic basement in the core of the ibero-Armorican Arc:
Ar-40/Ar-39 detrital mica ages complement the zircon's tale. Geology, 33(8):
637-640.
Harrington, H.J. and Kay, M., 1951. Cambrian and Ordovician faunas of eastern
Colombia. Journal of Paleontology, 25(5): 655-668.
Hoffman, P., 1991. Did the breakout of Laurentia turn Gondwanaland inside-out.
Science, 252(5011): 1409-1412.
Hoskin, P.W.O. and Black, L.P., 2000. Metamorphic zircon formation by solid-state
recrystallization of protolith igneous zircon. Journal of Metamorphic Geology,
18(4): 423-439.
Jimenez Mejia, D.M., Juliani, C. and Cordani, U.G., 2006. P-T-t conditions of high-grade
metamorphic rocks of the Garzon Massif, Andean basement, SE Colombia. J S
Am Earth Sci, 21(4): 322-336.
Johansson, A., 2009. Baltica, Amazonia and the SAMBA connection-1000 million years
of neighbourhood during the Proterozoic? Precambrian Res, 175(1-4): 221-234.
Johansson, L., Moller, C. and Soderlund, U., 2001. Geochronology of eclogite facies
metamorphism in the Sveconorwegian Province of SW Sweden. Precambrian
Res, 106(3-4): 261-275.
Johnston, S., Gehrels, G., Valencia, V. and Ruiz, J., 2009. Small-volume U-Pb zircon
geochronology by laser ablation-multicollector-ICP-MS. Chemical Geology,
259(3-4): 218-229.
Kemp, A.I.S. and Gray, C.M., 1999. Geological context of crustal anatexis and granitic
magmatism in the northeastern Glenelg River Complex, western Victoria.
Australian Journal of Earth Sciences, 46(3): 407-420.
Keppie, J., Dostal, J., Ortega-Gutierrez, F. and Lopez, R., 2001. A Grenvillian arc on the
margin of Amazonia: evidence from the southern Oaxacan Complex, southern
Mexico. Precambrian Res, 112(3-4): 165-181.
51
Keppie, J.D. and Dostal, J., 2007. Rift-related basalts in the 1.2-1.3 Ga granulites of the
northern Oaxacan Complex, southern Mexico: evidence for a rifted arc on the
northwestern margin of Amazonia. P Geologist Assoc, 118: 63-74.
Keppie, J.D. and Ortega-Gutierrez, F., 2010. 1.3-0.9 Ga Oaxaquia (Mexico): Remnant of
an arc/backarc on the northern margin of Amazonia. J S Am Earth Sci, 29(1): 2127.
Keppie, J.D. and Ramos, V.A., 1999. Odyssey of terranes in the Iapetus and Rheic oceans
during the Paleozoic. In: J.D. Keppie and V.A. Ramos (Editors), LaurentiaGondwana Connections Before Pangea. Geological Society of America Special
Paper, Boulder, CO.
Kovach, A., Fairbairn, H., Hurley, P., Basei, M. and Cordani, U., 1976. Reconnaissance
geochronology of basement rocks from Amazonas and Maranhão basins in Brazil.
Precambrian Res, 3(5): 471-480.
Kroonenberg, S., 1982. A Grenvillian granulite belt in the Colombian Andes and its
relation to the Guiana shield. Geol Mijnbouw, 61(4): 325-333.
Lawlor, P. et al., 1999. U-Pb geochronology, geochemistry, and provenance of the
Grenvillian Huiznopala Gneiss of Eastern Mexico. Precambrian Research, 94(12): 73-99.
Li, Z.X. et al., 2008. Assembly, configuration, and break-up history of Rodinia: A
synthesis. Precambrian Res, 160(1-2): 179-210.
Litherland, M. et al., 1989. The Proterozoic of eastern Bolivia and its relationship to the
Andean mobile belt. Precambrian Res, 43(3): 157-174.
LITHERLAND, M. and BLOOMFIELD, K., 1981. The Proterozoic history of eastern
Bolivia. Precambrian Res, 15(2): 157-&.
Loewy, S., Connelly, J., Dalziel, I. and Gower, C., 2003. Eastern Laurentia in Rodinia:
constraints from whole-rock Pb and U/Pb geochronology. Tectonophysics, 375(14): 169-197.
Ludwig, K., 2003. User's manual for Isoplot 3.00. A geochronological toolkit for
Microsoft Excel, 4. Berkeley Geochronology Center
Special Publication.
MacDonald, W.D. and Hurley, P.M., 1969. Precambrian gneisses from northern
Colombia, South America. Geological Society of America Bulletin, 80: 18671872.
52
McCourt, W., Aspden, J. and Brook, M., 1984. New geological and geochronological
data from the Colombian Andes - Continental growth by multiple accretion. J
Geol Soc London, 141(SEP): 831-845.
Mojica, J., Villaroel, C. and Macia, C., 1987. Nuevos afloramientos fosil ́ıferos del Ordov
́ıcico Medio (Fm El Higado) al oeste de Tarqui, valle superiordelMagdalena
(Huila,Colombia). Geologıa Colombiana, 16: 95-97.
Mosher, S., 1998. Tectonic evolution of the southern Laurentian Grenville orogenic belt.
Geological Society of America Bulletin, 110(11): 1357-1375.
Mosher, S., Levine, J.S.F. and Carlson, W.D., 2008. Mesoproterozoic plate tectonics: A
collisional model for the Grenville-aged orogenic belt in the Llano uplift, central
Texas. Geology, 36(1): 55-58.
Nance, R.D. et al., 2007. Neoproterozoic-early Paleozoic paleogeography of the periGondwanan terranes: Amazonian versus West African connections. In: E. Nasser
and J.-P. Liégeois (Editors), The Boundaries of the West African Craton.
Geological Society of London, Special Publication, pp. 345-383.
Nemchin, A. and Cawood, P., 2005. Discordance of the U-Pb system in detrital zircons:
Implication for provenance studies of sedimentary rocks. Sediment Geol, 182(14): 143-162.
Ordonez-Carmona, O., Pimentel, M.M. and De Moraes, R., 2002. Granulitas de Los
Mangos: un fragmento grenviliano en la parte SE de la Sierra Nevada de Santa
Marta. Revista de la Academia Colombiana de Ciencias Exactas, Fısicas y
Naturales, 26(99): 169-179.
Ordonez-Carmona, O., Pimentel, M.M., De Moraes, R. and Restrepo, J.J., 1999. Rocas
Grenvillianas en la region de Puerto Berrio – Antioquia. Revista de la Academia
Colombiana de Ciencias Exactas, F ́ısicas y Naturales, 23(87): 225-232.
Ordonez-Carmona, O., Restrepo Alvarez, J.J. and Pimentel, M.M., 2006.
Geochronological and isotopical review of pre-Devonian crustal basement of the
Colombian Andes. J S Am Earth Sci, 21(4): 372-382.
Parra, M. et al., 2009. Orogenic wedge advance in the northern Andes: Evidence from the
Oligocene-Miocene sedimentary record of the Medina Basin, Eastern Cordillera,
Colombia. Geological Society of America Bulletin, 121(5-6): 780-800.
Passchier, C.W. and Trouw, R.A.J., 2005. Microtectonics. Springer-Verlag, Berlin, 366
pp.
53
Pisarevsky, S.A., Wingate, M.T.D., Powell, C.M., Johnson, S. and Evans, D.A.D., 2003.
Models of Rodinia assembly and fragmentation. Geological Society, London,
Special Publications, 206(1): 35-55.
Priem, H. et al., 1982. Geochronology of the Precambrian in the Amazonas region of
southeastern Colombia (Western Guiana Shield). Geol Mijnbouw, 61(3): 229242.
Priem, H., Kroonenberg, S., Boelrijk, N. and Hebeda, E., 1989. Rb-Sr and K-Ar evidence
for the presence of a 1.6 Ga basement underlying the 1.2 Ga Garzon-Santa Marta
granulite belt in the Colombian Andes. Precambrian Res, 42(3-4): 315-324.
PRORADAM, 1979. La Amazonia Colombiana y sus recursos - Proyecto
Radargrametrico del Amazonas. Ingeominas, Bogota, 590 pp.
Ramos, V.A., 2010. The Grenville-age basement of the Andes. J S Am Earth Sci, 29(1):
77-91.
Reese, J. and Mosher, S., 2004. Kinematic constraints on Rodinia reconstructions from
the core of the Texas Grenville orogen. Journal of Geology, 112(2): 185-205.
Restrepo-Pace, P., Ruiz, J., Gehrels, G. and Cosca, M., 1997. Geochronology and Nd
isotopic data of Grenville-age rocks in the Colombian Andes: new constraints for
late Proterozoic Early Paleozoic paleocontinental reconstructions of the Americas.
Earth Planet Sc Lett, 150(3-4): 427-441.
Rivers, T., 1997. Lithotectonic elements of the Grenville Province: review and tectonic
implications. Precambrian Res, 86(3-4): 117-154.
Ruiz, J., Tosdal, R.M., Restrepo-Pace, P.A. and Murillo-Muñeton, G. (Editors), 1999. Pb
Isotope evidence for Colombian-southern Mexico connections in the Proterozoic.
GSA Special Paper, 336.
Sadowski, G. and Bettencourt, J., 1996. Mesoproterozoic tectonic correlations between
eastern Laurentia and the western border of the Amazon Craton. Precambrian
Res, 76(3-4): 213-227.
Santos, J. et al., 2000. A new understanding of the provinces of the Amazon craton based
on integration of field mapping and U-Pb and Sm-Nd geochronology. Gondwana
Research, 3(4): 453-488.
Santos, J. et al., 2003. Age, source, and regional stratigraphy of the Roraima Supergroup
and Roraima-like outliers in northern South America based on U-Pb
geochronology. Geol Soc Am Bull, 115(3): 331-348.
54
Santos, J.O.S. et al., 2008. Age and autochthonous evolution of the Sunsas Orogen in
West Amazon Craton based on mapping and U-Pb geochronology. Precambrian
Res, 165(3-4): 120-152.
Soderlund, U. et al., 2005. U-Pb baddeleyite ages and Hf, Nd isotope chemistry
constraining repeated mafic magmatism in the Fennoscandian Shield from 1.6 to
0.9 Ga. Contributions to Minralogy and Petrology, 150(2): 174-194.
Solari, L. et al., 2003. 990 and 1100 Ma Grenvillian tectonothermal events in the northern
Oaxacan Complex, southern Mexico: roots of an orogen. Tectonophysics, 365(14): 257-282.
Tassinari, C. and Macambira, M., 1999. Geochronological provinces of the Amazonian
Craton. Episodes, 22(3): 174-182.
Tassinari, C.C.G. et al., 1996. Geochronological systematics on basement rocks from the
Rio Negro–Juruena Province (Amazonian Craton), and tectonic implications.
International Geology Review, 38(2): 161-175.
Teixeira, W. et al., 2010. A review of the tectonic evolution of the Sunsás belt, SW
Amazonian Craton. J S Am Earth Sci, 29(1): 47-60.
Teixeira, W., Tassinari, C., Cordani, U. and Kawashita, K., 1989. A review of the
geochronology of the Amazonian Craton - Tectonic implications. Precambrian
Res, 42(3-4): 213-227.
Tohver, E. et al., 2004a. Terrane transfer during the Grenville orogeny: tracing the
Amazonian ancestry of southern Appalachian basement through Pb and Nd
isotopes. Earth Planet Sc Lett, 228(1-2): 161-176.
Tohver, E. et al., 2006. Restored transect across the exhumed Grenville orogen of
Laurentia and Amazonia, with implications for crustal architecture. Geology,
34(8): 669-672.
Tohver, E. et al., 2004b. Significance of the Nova Brasilandia metasedimentary belt in
western Brazil: Redefining the Mesoproterozoic boundary of the Amazon craton.
Tectonics, 23(6): TC6004.
Tohver, E., van der Pluijm, B., Van der Voo, R., Rizzotto, G. and Scandolara, J., 2002.
Paleogeography of the Amazon craton at 1.2 Ga: early Grenvillian collision with
the Llano segment of Laurentia. Earth Planet Sc Lett, 199(1-2): 185-200.
Toussaint, J.F., 1993. Precámbrico–Paleozóico., Evolución geológica de Colombia.
Universidad Nacional de Colombia, Medellin, pp. 129.
55
Trumpy, D., 1943. Pre-Cretaceous of Colombia. Geological Society American Bulletin,
54: 1281-1304.
Tschanz, C., Richard, F.M., Cruz, B.J., Harald, H.M. and Gerald, T.C., 1974. Geologic
evolution of the Sierra Nevada de Santa Marta, Northeastern Colombia.
Geological Society American Bulletin, 85: 273–284.
Vasquez, M., Altenberger, U. and Romer, R.L., 2009. Neogene magmatism and its
possible causal relationship with hydrocarbon generation in SW Colombia. Int J
Earth Sci, 98(5): 1053-1062.
Vernon, R.H., 2004. A Practical Guide to Rock Microstructure. Cambridge University
Press, 606 pp.
Walker, N., 1992. Middle Proterozoic geologic evolution of Llano Uplift, Texas Evidence from U-Pb zircon geochronometry. Geol Soc Am Bull, 104(4): 494504.
Waters, D., 2001. The significance of prograde and retrograde quartz-bearing intergrowth
microstructures in partially melted granulite-facies rocks. Lithos, 56(1): 97-110.
Weber, B. and Hecht, L., 2003. Petrology and geochemistry of metaigneous rocks from a
Grenvillian basement fragment in the Maya block: the Guichicovi complex,
Oaxaca, southern Mexico. Precambrian Res, 124(1): 41-67.
Weber, B. and Kohler, H., 1999. Sm-Nd, Rb-Sr and U-Pb geochronology of a Grenville
terrane in southern Mexico: origin and geologic history of the Guichicovi
complex. Precambrian Res, 96(3-4): 245-262.
Weber, B., Valencia, V.A., Schaaf, P., Pompa-Mera, V. and Ruiz, J., 2008. Significance
of Provenance Ages from the Chiapas Massif Complex (Southeastern Mexico):
Redefining the Paleozoic Basement of the Maya Block and Its Evolution in a
Peri-Gondwanan Realm. Journal of Geology, 116(6): 619-639.
56
FIGURE 1
Santa
Marta
Guajira
Northern Andes
foreland basins
Santander
San
Lucas
Macarena
Other wells referenced
Am
azo
4
3
Wells reported in this article
nC
CA
Garzon
MI
Fig.2
2
Amazon
on
Sao Luis
Craton
Graben
1
RNJ
Southern
Peru
rat
CA
VT
Parnaiba
Sao
Francisco
Craton
RSI
S
ne
ma
Arequipa-Antofalla
terrane
Central Amazonian
>2.5 Ga
MI
Maroni-Itacaiunas
2.2-1.95 Ga
VT
Ventuari-Tapajos
1.95-1.8 Ga
Rio Negro-Juruena
RNJ 1.8-1.55 Ga
RSI
S
Rondonia-Sn Ignacio
1.5-1.3 Ga
Sunsas - Aguapeí
1.3-0.9 Ga
na
ra
Pa
CA
pa
Amazonia
Geochron. Provinces
Pa
Cu
Luis Alves
Craton
Rio de la
Plata
Andean deformation front
Inferred cratonic limits and other
major structural features
pC massifs exposed in the Andes
Other pre-Pan African cratonic
blocks
Pan African - Brasiliano belts
0.9-0.6 Ga
Figure 2
Pa
c
ifi
c
a)
Early Mesoproterozoic and
Paleoproterozoic crystalline basement
North Andean Orogen
Late Mesoproterozoic (Grenville-age)
Cordilleran Inliers
Llanos Basin
Supracrustal metasedimentary units
of Mesoproterozoic age
Lower Paleozoic clastic sequences
(Araracuara Formation)
Macarena-2
ar
ej
Gü
Payara-1
-A
pa
Solita-1
d
te
)P
(?
Putumayo
Basin
b
ra
PR-3215
Thin Tertiary
cover
J-263
.
sF
oF
.
Pi
ta
Representative foliations Las
Minas gneisses and migmatites
lF
.
Magdalena
River valley
El Pital
Ag
ra
d
MIVS-37a
MIVS-41
CB-006
Ks
AH-1419
g
Al
GARZON
na
le
da
ag
M
m
r
He
.
sF
ra
i
ec
Tarqui
Altamira
Guadalupe
MIVS-11
r
R
ive
Lower/Upper tertiary sandstones
and aluvial deposits
Cretaceous Clastic sequences
Jurassic Vulcanosedimentary
El Vergel Granulites
Las Margaritas Migmatites
El Recreo Gneiss
Guapoton-Mancagua
orthogneiss
Las Minas Migmatites
El Recreo
Jurassic Granitoids
CB-002
Permo-Triassic Granitoids
MIVS-26
0
a
Js
a
os
Quaternary aluvial
Garzón Group
La Plata
0°
180°
CJR-19
PROTEROZOIC UNITS
Serrania de
Las Minas
n=8
Brazil
PR-3092
Foreland Basins
b)
en
Ecuador
zG
Mandur-3
r
ve
in
Caiman-3
AH-1216
ris
Mandur-2
po
La Rastra-1
Garzon Massif
5
Km
10
Cambro-Ordovician clastic
sequences
Municipalities
Rivers
Roads
FIGURE 3
a)
data-point error ellipses are 68.3% conf.
0.30
MIVS-11
1550
“El Vergel Granulites”
Metamorphic Age
MSWD = 0.32
0.26
Metamorphic Overgrowths
900
990 ± 12 Ma (1.2%)
206
MSWD = 0.58
1.8
2.2
2.6
3.0
100μm
3.4
Mean 206/207 (2σ) = 1005 ± 26
MSWD = 0.19
0.13
3.8
1.2
d) 0.24
CB-002
Qz-Feldspar Gneiss
0.22
Pb/238U
MSWD = 1.08
0.20
18μm spots, Ion-Counters,
Metamorphic overgrowths
1.8
40μm, Faradays, >5% Disc.
and mixed analyses
100μm
2.8
3.8
1.4
972 ± 14 Ma (1.5%)
0.20
Mean 206/207 (2σ) = 977 ± 25
MSWD = 0.057
0.18
1050
1.8
2.2
2.6
3.0
.
f ) 0.30
0.22
100μm
0.14
CB-006
MIVS-37a
1700
Metasedimentary
Migmatitic Gneiss
(Paleosome)
1350
1500
0.26
1250
MSWD = 0.14
Pb/238U
1200
40μm, Faradays, <5% Disc.
Metamorphic Overgrowths
Pb/238U
1150
1300
0.22
1100
206
206
MSWD = 0.97
1100
206
30μm spots, Metamorphic rims
40μm, Detrital cores, <5% Disc.
40μm, Detrital cores, >5% Disc.
Biotite Gneiss
0.18
950
0.16
1400
Mean 206/207 (2σ) = 997 ± 14
MSWD = 0.04
0.16
800
0.26
0.24
1326 ± 13 Ma (1.0%)
2.4
Metamorphic Overgrowths
1000
e)
2.2
0.18
1000
0.10
0.8
2.0
Igneous Protolith Age
991 ± 12 Ma (1.2%)
1200
0.14
1.8
MIVS-41
1400
Mean 206/207 (2σ) = 998 ± 18
MSWD = 0.082
0.18
1.6
100μm
Mean 206/207 (2σ) = 1335 ± 13
MSWD = 0.53
Metamorphic Overgrowths
992 ± 12 Ma (1.2%)
1.4
Las Minas Augen-Gneiss
1600
206
Pb/238U
1000
0.15
MSWD = 0.45
0.22
0.17
30μm spots, Metamophic rims
0.14
1.4
0.30
1100
Pb/238U
Pb/238U
206
1050
40μm, Detrital cores, <5% Disc.
40μm, Detrital cores, >5% Disc.
1200
Mean 206/207 (2σ) = 1129 ± 12
MSWD = 0.49
0.19
1150
c)
1135 ± 12 Ma (1.1%)
MSWD = 0.96
1250
950
40μm, Faradays, Cores and rims
40μm, Faradays, Mixed ablation spots
Igneous Protolith Age
0.21
1350
Mean 206/207 (2σ) = 995 ± 5
MSWD = 0.22
0.18
MIVS-26
1450
992 ± 10 Ma (1.01%)
0.22
data-point error ellipses are 68.3% conf.
Guapoton Augen-Gneiss
Felsic Granulite
0.26
b) 0.23
900
0.14
0.14
0.12
30μm spots, Metamorphic rims
40μm, Detrital cores, <5% Disc.
40μm, Detrital cores, >5% Disc.
1.4
1.8
207
2.2
Pb/235U
30μm, Faradays, <5% Disc.
30μm, Faradays, >5% Disc.
700
100μm
2.6
0.10
3.0
0.5
1.5
207
2.5
Pb/235U
100μm
3.5
4.5
FIGURE 4
0.29
0.27
data-point error ellipses are 68.3% conf.
Macarena-2
Melanocratic mylonitic gneiss
1550
Igneous Protolith Age
1461 ± 15 Ma (1.0%)
Pb/238U
MSWD = 0.37
Mean 206/207 (2σ) = 1471 ± 18
MSWD = 0.17
1450
206
0.25
0.23
1350
40μm, Faradays, <5% Disc.
40μm, Faradays, >5% Disc.
0.21
2.5
2.7
2.9
3.1
207
3.3
Pb/235U
100μm
3.5
3.7
3.9
FIGURE 5
A
B
C
D
FIGURE 6
Payara-1 well
MSWD = 0.97
0.28
Pb/238U
Pb/238U
c
ce
1200
986 ± 19 Ma (1.9%)
2
1150
100μm
3
0.0
4
Mandur-2 well
0
d)
1800
Melanocratic Migmatitic Gneiss
0.19
0.18
Pb/238U
Pb/238U
Mean 206/207 (2σ) = 1587 ± 18
MSWD = 0.67
206
1400
0.22
Upper Intercept
Leucocratic dike
1017 ± 10 Ma (0.9%)
2
Migmatite
1800
0.32
10μm spots, Ion-Counters
30μm, Faradays, <5% Disc.
30μm, Faradays, >5% Disc.
1600
Metamorphic Overgrowths
800
1.5
2.5
207
3.5
235
10μm spots, Ion-Counters
30μm, Faradays, <5% Disc.
30μm, Faradays, >5% Disc.
Pb/
U
6.5
0.08
0.5
2.05
1800
1600
1200
1000
Crystallization Age
952 ± 21 Ma (2.2%)
MSWD = 0.98
100μm
Mean 206/207 (2σ) = 950 ± 25
MSWD = 0.91
100μm
5.5
1.95
1400
0.12
4.5
1.85
2000
Leucogranite
800
MSWD = 0.7
Mean 206/207 (2σ) = 991 ± 13
MSWD = 1.05
1.75
sample-2
0.20
0.16
989 ± 14 Ma (1.4%)
0.12
1.65
Caiman-3 well
Pb/238U
Pb/238U
1000
1.55
0.24
1200
0.20
100μm
0.28
1400
0.24
206
f ) 0.36
2000
940
0.15
1.45
5
sample-1
0.28
0.08
0.5
4
Caiman-3 well
0.32
0.16
3
100μm
206
e) 0.36
1060
MSWD = 0.66
0.17
MSWD = 0.62
Mean 206/207 (2σ) = 1019 ± 20
MSWD = 0.23
1
5
Mean 206/207 (2σ) = 1014 ± 12
MSWD = 0.24
0.16
1019 ± 23 Ma (2.2%)
0.14
4
980
Metamorphic Overgrowths
1000
3
Mandur-2 well
1200
0.18
2
Crystallization Age
MSWD = 0.62
0.26
1
100μm
<20% Disc.
950
1100
Protolith Age (Upp. Intercept)
1601 ± 28 Ma (1.7%) 1600
206
box heights are 1σ
600
1050
0.34
0.30
1400
1000
MSWD = 0.49
Mean 206/207 (2σ) = 985 ± 23
MSWD = 0.70
(6 analyses)
1
Ma
Upper Discordia Intercept
1046 ± 25 Ma (2.5%)
MSWD = 0.54
0.2
0.1
Metamorphic Overgrowths
800
000
206
206
d
or
1800
~2
to
Mean 206/207 Age (2σ)
1044 ± 26 Ma (2.5%)
MSWD = 0.5
ren
e
ref
<30% Discordant
>30% Discordant
Inheritance and mixing
Metamorphic Overgrowths
h
0.16
c)
0.3
1400
0.24
1000
Solita-1 well
Sheared Migmatite
Mean 206/207 (2σ) = 1608 ± 15
MSWD = 0.93
18μm spots, Ion-Counters
40μm, Faradays, <5% Disc.
40μm, Faradays, >5% Disc.
0.20
data-point error ellipses are 68.3% conf.
1700
1606 ± 16 Ma (1.0%)
Migmatitic Gneiss
0.12
b)
data-point error ellipses are 68.3% conf.
Igneous Protolith Age
206/207 AGE
a) 0.32
1.5
2.5
207
3.5
Pb/235U
4.5
5.5
6.5
FIGURE 7
a)
data-point error ellipses are 68.3% conf.
PR-3215 (Araracuara)
0.34
b) 0.36
Weakly sheared Syenogranite
1756 ± 18 Ma (1.0%)
Crystallization Age
1700
MSWD = 0.98
1700
Mean 206/207 (2σ), 16 analyses
1732 ± 24 Ma (1.4%)
0.28
1500
1600
MSWD = 0.66
Pb/238U
Mean 206/207 (2σ) = 1750 ± 18
MSWD = 0.43
(38 analyses)
206
Pb/238U
206
1800
0.32
Crystallization Age
0.26
35μm, Faradays, <10% Disc.
35μm, Faradays, >10% Disc.
J-263 (Araracuara)
1900
Deformed syenogranite
0.30
data-point error ellipses are 68.3% conf.
1500
1400
0.24
1300
c)
1100
1.5
2.5
3.5
4.5
1900
PR-3092 (Mitu)
MSWD = 0.76
Pb/238U
Pb/238U
0.35
e)
206
4
0.36
2100
1530 ± 24 Ma (1.6%)
2
Pb/238U
206
206
box heights are 1
1520
2
3
207
235
Pb/
U
5
6
7
10
12
14
1900
Crystallization Age
MSWD = 1.03
0.32
1800
1700
0.30
0.24
2000
35 analyses, Mean 206/207 (2σ)
box heights are 1
1600
1640
1500
0.26
4
8
CJR-19 (Taraira)
1440
0.20
6
Undeformed Granitoid
0.28
1600
206/207 Age
1300
4
0.38
0.28
0.24
1580
1588 ± 21 Ma (1.3%)
1700
1500
1660
1500
0
0.34
MSWD = 0.32
0.32
800
0.36
1900
11 analyses, Mean 206/207 (2σ)
Pb/238U
0.05
f)
Crystallization
Inheritance
Crystallization Age
box heights are 1
1740
100μm
6
2000
1600
0.15
5
AH-1419
Taraira Granitoid
0.40
2400
0.25
1200
MSWD = 0.37
Mean 206/207 (2σ) = 1574 ± 17
MSWD = 0.20
(27 analyses)
3
Crystalization, <10% Disc.
Crystallization Age
1574 ± 16 Ma (1.0%)
2
5
Inheritance, <30% Discordant
>30% Discordant analyses
Undeformed granite
Crystallization Age
0.18
4
1578 ± 30 Ma (1.9%)
1500
1300
1660
Mean 206/207 (2σ), 9 analyses
1700
0.22
1740
3
0.45
35μm, Faradays, <5% Disc.
35μm, Faradays, >5% Disc.
Inheritance and mixing
0.26
2
d)
AH-1216 (Mitu)
Undeformed Granitoid
0.30
206
0.16
5.5
0.38
0.34
206/207 Age
35μm, Faradays, <5% Disc.
35μm, Faradays, >5% Disc.
206/207 Age
0.18
0.20
206/207 Age
0.22
box heights are 1
1820
1300
3
4
1560
1480
207
5
Pb/235U
6
7
FIGURE 8
O
O-N-H-Gu
G
Mexico
Mi
Ma MT P
Ar
U-Pb age of
Metamorphism/
Anatexis
Colombia
JRG-20-96
Dibulla Gneiss
All data (59)
<5% Disc. (41)
991±12 Ma
Santa Marta
Massif
Periods of relevant
Igneous Activity
972±14 Ma
MDA
~1086±34 Ma
MIVS-37a
Pital Migmatites
All data (70)
<5% Disc. (55)
N.A.
Las Minas Massif
CB-006
El Zancudo Migmatites
All data (111)
<5% Disc. (102)
MDA
~1150±18 Ma
CB-002
992±12 Ma
MDA
~1013±11Ma
MIVS-11
El Vergel Granulites
All data (89)
<5% Disc. (62)
992±10 Ma
Garzón Massif
El Vergel Granulites
All data (89)
<5% Disc. (59)
MDA
~1115±10 Ma
Leucogranite
All data (84)
<5% Disc. (61)
952±21 Ma
Caiman-3 Well
MDA
~1444±9 Ma
Migmatite
All data (80)
<5% Disc. (63)
989±14 Ma
Mitu-Taraira-Araracuara
Craton Granitoids
All data (278)
<5% Disc. (168)
Amazonia Provinces
Sunsas-Putumayo
1000
1200
RSI
1400
RNJ
1600
Age (Ma)
Foreland “in-situ” Amazonia Basement
Caiman-3 Well
VTP
1800
2000
FIGURE 9
NW Amazonia
N Andes Foreland Basement
North Andean Inliers
(Autochtonous NW Gondwana)
SW Amazonia
(para-autochtonous terranes)
Mexican Terranes
Regional cooling and
localized Low-P Metamorph
Millions of Years (Ma)
Margaritas Migmatites (Garzon)
Sedimentation
?
(Arc-related basins)
?
1100
Back-arc
sedimentation over
the continental
margin (?)
?
1200
This work
Arc Magmatism Metamorphic
event. Zr rims
Garzon Massif Dibulla gneiss
Metamorphic
event. Zr rims
Jojoncito gneiss
Sedimentation
This Work
Cardona et al. (2010)
Ordonez-Carmona et al (2007)
Cordani et al. (2005)
Restrepo-Pace et al. (1997)
Post-tectonic pegmatites
Rondônia tin province
granites followed by
regional cooling
Peak Metamorphism
“Zapotecan” event
AMCG magmatism
Strike-slip continuous deform.
(Late shearing offshots and final
accretion of the “Paragua block”)
Post-collisional
Sunsas Granite Suite
Sunsas group
collisional deformation
?
“Olmecan” event
Passive-margin and
intra-continental rift
sedimentation
Boger et al. (2005)
Santos et al. (2008)
Cordani et al. (2009)
Teixeira et al. (2010)
Tohver et al. (2004)
Laurentia NE
(Grenville)
Dalane phase
(Post-collisional
relaxation and collapse)
Extensional
Phase
Falkenberg phase
Agder phase
(Main continental
collision event)
Post-tectonic
granites
Arendal phase
Early terrane collisions
Arc/Backarc magmatism
(Protolith ages in orthogneisses.
Huiznopala, Oaxaca, Novillo)
Laurentia SE
(Llano)
Bimodal magmatism
and Sedimentation
Accretion of
older than
1.25 Ga terranes
Grenville Orogen
Fringing arc /terrane
collisions initiation (?)
(Migmatization
Mandur and Solita wells)
Peak Metamorphism
Garzon and Las Minas
Sunsas Aguapei Orogen
1000
Migmatization
Payara and Caiman wells
Putumayo Orogen
Undeformed Leucogranites
SW Baltica
Sveconorwegian Orogen
900
Rigolet Phase
Ottawan
Phase
Adirondian
Magmatism
(AMCG suites)
Shawinigan
phase
Keppie et al. (2009)
Cameron et al. (2004)
Keppie et al. (2003)
Solari et al. (2003)
Bingen et al. (2008)a
Bingen et al. (2008)b
Bogdanova et al. (2008)
Barker and Reed
(2010)
Mosher et al. (2004)
Reese and Mosher
(2004)
Bartholomew et al.
(2010)
Tollo et al. (2004)
Gower and Krogh
(2002)
Rivers (1997)
McLelland (1996)
FIGURE 10
a)
b)
pre-1.25 Ga Baltica
1.07-1.01 Ga
Newborn Asgard Sea
~1.14 Ga
Accretion of
non-Amazonian Telemarkia
(Arendal phase)
Fennoscandia
Idefjorden
Arc accretion onto
Amazonia’s margin.
Metamorphism and melting
under Amphibolite facies
conditions.
Telemarkia
?
Laurentia
“Oaxaquian-Colombian terranes”
Fringing arc system
outboard of Amazonia
A
?
PF
?
GR
BR
APC
Backarc spreading and
sedimentation over
the continental margin
P
P
LLA
~1.2 Ga
~1.15 Ga
Metamorphism
Nova Brasilandia belt
RSI
1000 km
1000 km
Amazonia
Baltica-Amazonia collision initiation close to their final Rodinia positions.
Amalgamation of fringing arcs and terranes.
Laurentia-Amazonia initial collision.
Development of the “Colombian-Oaxaquian”fringing-arc system.
c)
1.00 - 0.98 Ga
Val
h
alla
Archean crust
Paleoproterozoic
Asgard Sea
Oro
gen
nide
a
Tim
Baltica
gin
Mar
Laurentia
Idefjorden terrane
Grenville Province
Telemarkia terrane
Granite-Rhyolite province
?
Gande
rian
massif
s
Eastern Segment
Blekinge-Dalarna
dolerite dike swarm
Avalonian Cadomian
terranes
Amazonia
Rondonia-San Ignacio
Colombian-Oaxaquian
terranes
Sunsas-Aguapei
Putumayo basin basement
West-Africa
Nova Brasilandia
cooled below ~500°C
1000 km
Rodinia fully assembled
Maximum convergence along the Sveconorwegian-Putumayo orogen
A= Adirondacks
APC = Argentine Precordillera Terrane
BR = Blue Ridge
GR = Granite-Rhyolite
LLA = Llano Segment
P = Paragua Terrane
PF = Paleoproterozoic underlying the Andean foreland
RSI = Rondonia-San Ignacio Orogen
Other blocks were omitted for simplicity. Craton outlines and general geology: Laurentia, Cawood et al. 2007.
Baltica, Bogdanova et al. 2008, Bingen et al. 2008, Amazonia, Cordani et al. 2001
FIGURE 11
“Oaxaquia-Colombian terranes”
peri-Amazonian fringing arc
@ ~1.15 Ga
Detritus deposited on Amazonia’s
Paleoproterozoic basement
(Cratonic provenance signatures with little
or no arc influence)
Arc- related volcanosedimentary
sequences (”Garzon group”-like)
Mildly extensive
back-arc
Sea Level
Oceanic Crust
?
Intrusion of arc plutons with
variable older crustal contamination
(Guapoton and Oaxaquia
orthogneisses)
Mid- Mesoproterozoic to
Paleoproterozoic basement
of the RNJ province