In the Gurupi region, located at the border between the Pará and Maranhão states in
northern Brazil, igneous and metamorphic rocks crop out as part of the Parnaíba Structural
Province. Early geochronological studies, based on the Rb-Sr and K-Ar methods have shown two
geochronological domains. The rocks that crop out towards the Atlantic margin showed a
Paleoproterozoic signature, around 2000 Ma, whereas the rocks that crop out towards the inner
portions of the continent showed a Neoproterozoic signature, especially between 800 and 500
Ma. These domains have been then defined as the São Luís Craton and Gurupi Belt, respectively.
Several lithostratigraphic propositions have been developed throughout more than two decades.
However, these propositions always lack robust geochronological support. Geotectonic models
discussed a one- or two-phase evolution for the Gurupi Belt, also lacking robust geochronological
and isotopic data to consubstantiate the interpretations. Furthermore, among the several gold
deposits that occur in both the cratonic and belt areas, only a few have geological and genetic
information. These subjects are addressed in more or less depth by this thesis.
New propositions for the regional lithostratigraphy and geological evolution have been
achieved in this work by revaluating the available geological, geochemical, geochronological and
isotopic dataset, as well as by adding new geochronological data on zircon (Pb-evaporation, U-Pb
ID-TIMS, and LAM-ICP-MS) for most of the igneous and orthometamorphic rocks in the region.
Whole rock Nd isotope data have also been obtained, allowing the discussion of crustal accretion
and reworking. The results show a rather complex geological evolution with intensive and
extensive crustal growth between 2.24-2.15 Ga and crustal reworking, involving melting,
migmatization, metamorphism, and deformation around 2.10 Ga. The following results have been
obtained for the São Luís Craton: Aurizona Group, metavolcano-sedimentary sequence,
maximum age of 2241 Ma (juvenile) that possibly evolved until c.a. 2200 Ma; Tromaí Intrusive
Suite, calc-alkaline, metaluminous tonalites of oceanic island arc, 2168 Ma (juvenile); Areal
Granite, calc-alkaline, weakly peraluminous, 2150 Ma (mixing of juvenile and arc materials). In
the Gurupi Belt, the following results have been obtained: Igarapé Grande Metatonalite, small
and localized granoblastic tonalite, 2594 Ma; Itapeva Complex, weakly migmatized tonalitic
orthogneiss, 2167 Ma (mostly juvenile); Chega Tudo Formation, metavolcano-sedimentary
sequence (back-arc basin?), 2150-2160 Ma; Maria Suprema Granite, syntectonic, peraluminous
muscovite-bearing granite, 2100 Ma (similar to other peraluminous granitoids in the Gurupi
Belt). The Gurupi Group is tentatively placed in the Paleoproterozoic (>2160 Ma), but this must
still be proved. The above data are interpreted on a plate tectonics basis, as follows. An oceanic
basin is open at ca. 2260 Ma and is followed by the onset of subduction, formation of island arcs
and voluminous calc-alkaline magmatism in oceanic settings, and concomitant reworking of the
arcs between 2170-2150 Ma. This set of oceanic terranes has been accreted (soft-collision) onto
an Archean continental margin to southwest (Archean part of the Amazonian Craton or a present
day concealed cratonic nuclei). The collision provoked the metamorphism, deformation, and
partial melting of the newly formed Paleoproterozoic crust and of part of the Archean bloc, or
their erosive detritus, migmatization, and emplacement of peraluminous granitoids at 2100-2080
Ma. The region has been the locus of a second event in the Neoproterozoic. A continental rift
developed in the bloc that was assembled in the Paleoproterozoic, as attested by the intrusion of a
nepheline syenite (Boca Nova) at 732 Ma. Sedimentary rocks that filled this rift (Marajupema
Formation) have detrital zircon crystals that show the youngest ages around 1100 Ma. The rift
evolved probably to an oceanic basin, as suggested by the widespread occurrence of detrital
zircons with ages around 550 Ma in small sedimentary basins that have been filled with immature sediments. The precise time of orogenesis climax that followed basin closure, with mass transport
from SSW to NNE and accompanying metamorphism, is not yet constrained. Equivocal
geochronological information point to 650-520 Ma (zircon of the nepheline syenite, Rb-Sr and KAr
ages in minerals).
The metallogeny of selected gold deposits occurring in both the São Luís Craton and the
Gurupi Belt is addressed using varied information, such as geology, chlorite chemistry, fluid
inclusion geochemistry, and stable (O, H, C, S) and radiogenic (Pb) isotopes. Structural and
textural relationships, and Pb isotope data indicate a post metamorphic peak and late- to posttectonic
timing for the gold mineralization with respect to the Paleoproterozoic events (post 2080
Ma). At a regional scale, the deposits show a similar signature characterized by formation
temperatures between 280° and 380°C; pressures of 2-3 kbars; low-salinity (5 mass % NaCl
equiv), reduced and moderately dense aqueous-carbonic (CO2 <20 mol%, traces of CH4 and N2),
showing strong evidence for phase separation. Stable isotope studies suggest distinct sources for
fluids and solutes. The carbonate, graphite, and fluid inclusion carbon comes from two sources: a
depleted organic source, and an unknown source that may be magmatic, metamorphic or mantlederived
(or both). Sulfide sulfur derived directly from magmas or from the dissolution of
magmatic sulfides. Combined oxygen and hydrogen isotopes attest a metamorphic source for the
fluids. Therefore, dehydration and decarbonization reactions during the metamorphism of the
Paleoproterozoic metavolcano-sedimentary sequences appear to have produced the mineralizing
fluids. Gold was transported as a reduced sulfur complex, such as the Au(HS)2
- and precipitated
in response to the breakdown of this complex due to phase separation and fluid-rock interactions.
The geological and genetic constraints are consistent with the orogenic gold model, found in
metamorphic belts of all ages.
As a whole the results of this study have implications for the understanding of the
Paleoproterozoic and Neoproterozoic orogenies that built up the South American Platform and
for the assembly and break-up of the Atlantica, Rodinia, and West-Gondwana supercontinents.
The geological scenario outlined here for the Paleoproterozoic shows good correlations with
those found especially in the southeastern Guyana Shield and in the southern portion of the West-
African Craton. For the Neoproterozoic, the available information is still insufficient to draw
major correlations.
|
