- Email: [email protected]
Contribution of global potential field data to the tectonic reconstruction of the Rio Grande Rise in the South Atlantic
Accepted Manuscript Contribution of global potential field data to the tectonic reconstruction of the Rio Grande Rise in the South Atlantic Igor Leonardo Guerra Galvão, David Lopes de Castro PII:
S0264-8172(17)30255-6
DOI:
10.1016/j.marpetgeo.2017.06.048
Reference:
JMPG 2982
To appear in:
Marine and Petroleum Geology
Received Date: 11 March 2017 Revised Date:
26 May 2017
Accepted Date: 30 June 2017
Please cite this article as: Guerra Galvão, I.L., Lopes de Castro, D., Contribution of global potential field data to the tectonic reconstruction of the Rio Grande Rise in the South Atlantic, Marine and Petroleum Geology (2017), doi: 10.1016/j.marpetgeo.2017.06.048. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT 1
1
Contribution of Global Potential Field Data to the Tectonic Reconstruction of the Rio
2
Grande Rise in the South Atlantic
RI PT
3
Igor Leonardo Guerra Galvão, David Lopes de Castro
5
Programa de Pós-Graduação em Geodinâmica e Geofísica da Universidade Federal do Rio
6
Grande do Norte, Campus Universitário S/N, 59078-970, Natal, Brasil
7
Corresponding Author: David Lopes de Castro ([email protected])
SC
4
9
M AN U
8
Highlights
• The Rio Grande Rise (South Atlantic) was studied based on potential field data
11
• Regional gravity low indicates crustal thickness beneath the Rio Grande Rise
12
• Magnetic anomalies shed light on the Rio Grande Rise tectonic evolution
13
• Oceanic fracture zones control the development of the rise
14
• Simplified reconstruction model of the Rio Grande Rise is proposed and discussed
AC C
EP
TE D
10
ACCEPTED MANUSCRIPT 2
15
ABSTRACT
16
The application of advanced enhancement techniques for geophysical anomalies to global
RI PT
17
gravity (WGM2012) and magnetic (EMAG2) models sheds light on the complex tectonic
19
evolution of the Rio Grande Rise (RGR) in the southern South Atlantic. Long wavelength
20
Bouguer gravity lows indicate a thicker crust beneath of the ridge, whose nature can be related to
21
a microcontinent or an excess of volcanism within the oceanic realm. Recently dredged
22
continental rocks reinforce the hypothesis of a microcontinent or, at least, slivers of continental
23
crust. However, the reserval magnetic pattern of the processed magnetic anomalies provide no
24
evidence of aborted spreading center similar to the well-studied Jan Mayen microcontinent and
25
the surrounding (inactive) Aegir and (active) Kolbeinsey ridges in the North Atlantic Ocean. The
26
reversal magnetic anomalies show a series N-S trending parallel stripes roughly follow the
27
current South American coastline and segmented by E-W oriented oceanic fracture zones (FZs).
28
The magnetic stripes are bended eastwards at the RGR, showing a more complex magnetic
29
pattern similar to that in the Iceland. The aborted Cruzeiro do Sul Rift (CSR) and the Jean
30
Charcot Chain (JCC) are structures that cross the RGR and contribute to the understanding of the
31
tectonic evolution of the South Atlantic Ocean. NW-SE oriented extensive gravity lows and
32
bathymetric valleys, which mark the CSR, are segmented by E-W trending magnetic lineaments
33
related to FZs. This structural configuration suggests that the extensional event, which formed the
34
rift and the seamounts chain, was replaced by strike-slip movements along the FZs. In addition,
35
we constructed a plate kinematic model for the evolution of the RGR based on bathymetric, free-
36
air and Bouguer gravity and magnetic data. Our model comprises five main stages of the RGR
37
formation and evolution between late Cretaceous and Paleocene (ca. 95 - 60 Ma), separated by
AC C
EP
TE D
M AN U
SC
18
ACCEPTED MANUSCRIPT 3
published seafloor isochrones. The proposed model suggests that the RGR was built at the mid-
39
Atlantic ridge by increased magmatism probably related to the Tristan da Cunha hotspot.
40
Keywords: Potential Field; Aseismic Ridge; Tectonics; South Atlantic
RI PT
38
41 42
1. INTRODUCTION
SC
43
Geodynamic processes of continental break-up and seafloor spreading left behind various
45
large features throughout the world's oceans. These structures detach themselves from the normal
46
oceanic crust by forming large semi-circular or sometimes linear regions such as aseismic ridges
47
or rises. Kumar (1979) lists three possible origins for these features: (1) microcontinents, (2)
48
uplifted oceanic crust, and (3) linear volcanic features. However, this same author points to the
49
possible existence of complex structures, yielded by a combination of the tectonic mechanisms
50
mentioned above. The Rio Grande Rise (RGR) is an example of a large oceanic rise in the South
51
American plate, whose complex origin and tectonic evolution still a matter of debate. The RGR is
52
a mountain range that covers 3,000 km² and rises 3,500 m above the surrounding abyssal plain,
53
located 1,500 km off the Brazilian coast between 27°S and 36°S latitude and 27°W and 40°W
54
longitude (Figure 1). It separates the Brazil (North) and Argentina (South) oceanic basins and
55
represents an extensive barrier to sediment transport between these basins through seafloor
56
currents. The Vema and Hunter channels are the main connections between these oceanic basins
57
(Kaji et al., 2011).
58
AC C
EP
TE D
M AN U
44
The RGR is defined as a semicircular aseismic southern segment of the South Atlantic,
59
and his given name was associated with the absence of earthquakes (e.g., Gamboa and
60
Rabinowitz, 1984). This peculiar morphology, similar to other elevated areas, suggests that these
ACCEPTED MANUSCRIPT 4
regions have different geological histories than the adjacent oceanic crust. These regions
62
sometimes represent isolated fragments of continental crust on the seafloor, such as the Orphan
63
Knoll in the North Atlantic (Chian et al., 2001) and the Jan Mayen microcontinent in northern
64
Iceland (Kodaira et al., 1998). On the other hand, most of these aseismic rises, probably such as
65
the RGR (Gamboa and Rabinowitz, 1984), predominantly consist of basaltic oceanic rocks where
66
sampled.
SC
67
RI PT
61
The RGR formed within a series of structures related to the rifting and opening of the South Atlantic Ocean in the early Cretaceous (130 Ma) (Cobbold et al., 2001; Zálan and Oliveira,
69
2005; Ricommini, 2008). The transit of the lithospheric plate over the Tristan da Cunha Plume
70
probably caused the extensional events that resulted in the break-up of the Gondwana
71
supercontinent and resulted on extensive basaltic lava flows during the Neocomian, collectively
72
known as the Paraná-Etendeka large Igneous Province (Fodor et al., 1984; Mitzusaki, 1986;
73
O'Connor and Duncan, 1990). This plume was responsible for generating a thermal event that
74
reached the Western Gondwana continental lithosphere, particularly in southeastern Brazil, where
75
a wide area was uplifted and underwent changes in its rheology to become more ductile and less
76
resistant to stretching (Estrella, 1972; Asmus and Ferrari, 1978; Asmus, 1975, 1984; Ojeda, 1982;
77
Fulfaro et al., 1982; Macedo, 1989).
TE D
EP
AC C
78
M AN U
68
O'Connor and Duncan (1990) associated the formation of the RGR-Walvis Ridge system
79
to the presence of the Tristan da Cunha-Gough Plume, which is located on the spreading axis of
80
the Mid-Atlantic ridge with its large supply of magma. Gibson et al. (2005) noted that the mantle
81
plume is currently located 550 km from the central axis of the range between the Tristan da
82
Cunha and Gough islands. Recent work by Stica et al. (2014, their Fig. 14) considers the Rio
83
Grande Rise to be a continental block covered by thick volcanic deposits of unknown origin
ACCEPTED MANUSCRIPT 5
based on newly dredged rock samples of Precambrian granitic and metamorphic rocks in
85
oceanographic expeditions lead by the Brazilian Geological Survey (BSG) and a pool of
86
Brazilian and international universities in 2011 and 2013 (Fioravanti, 2014). Pushcharovsky
87
(2013) also pointed out that RGR is a microcontinent spatially related to the São Paulo
88
continental prominence and separated by the N-S oriented deep trough of the Vema Channel
89
(Figures 1 and 2).The presence of typically continental crust rocks on seabed, although literally
90
essential, is nevertheless insufficient to define the RGR as a continental fragment of the South
91
American plate. Other key evidences, such as extinct spreading ridge, and jumped spreading
92
center, are still necessary to confirm this new approach. As well, geophysical, geological and
93
geochronological datasets must be reinterpreted and plate tectonic reconstruction models must be
94
reviewed to accommodate an ancient microcontinent 1300 km far away of the current coast line.
95
However, the characteristics of the RGR make it difficult to precisely determine its genesis,
96
especially regarding its depth and the difficulty of processes occurring at such depths. Its
97
geotectonic evolution is still disputed, and information about its structural framework is scarce
98
and requires better explanations.
SC
M AN U
TE D
EP
100
In this study, we apply advanced enhancement techniques for gravity and magnetic
AC C
99
RI PT
84
101
anomalies to investigate the formation of the RGR from the interaction between the tectonic
102
processes of the evolution of both the Mid-Atlantic Ridge and the Tristan da Cunha hotspot
103
during the opening of the South Atlantic in a regional context. The structural control of the
104
oceanic fracture zones (FZs) and the extensional tectonic events associated with the evolution of
105
the extensive failed Cruzeiro do Sul Rift and the formation of the southern portion of the Jean
106
Charcot seamount chain are also examined. We show a reconstruction model of the RGR from
ACCEPTED MANUSCRIPT 6
100 Ma to 55 Ma based on bathymetric, gravity and magnetic data and discuss the origin of this
108
large aseismic rise concerning to crustal thickening, seafloor spreading and structural control by
109
fracture zones. In summary, the novelties of this paper are: 1) crustal thickness beneath the RGR
110
derived from regional Bouguer anomaly; 2) areal distribution of depocenters and seamounts in
111
the aseismic bulge region unraveled by residual Bouguer and free-air anomalies; 3) detailed
112
seafloor spreading and fracture zones mapped using enhanced magnetic anomalies; and 4)
113
detailed plate motion reconstruction of the RGR based on bathymetric, gravity and magnetic data
114
including discussions about the crustal nature of the rise, tectonics of the Cruzeiro do Sul rift and
115
the role of the fracture zones in the context of the South Atlantic opening.
M AN U
SC
RI PT
107
116 117
2. REGIONAL GEOLOGICAL-STRUCTURAL SETTING
118
120
2.1. Rio Grande Rise (RGR)
TE D
119
The RGR is located on the SE coast of South America (Figure 1) and significantly
122
modifies the planar morphology of the abyssal oceanic domain between the Brazil and Argentina
123
oceanic basins in the western South Atlantic (Adams, 1981). The Walvis Ridge, other important
124
tectonic element in the South Atlantic Ocean, is a NE-SW chain located on the opposite side of
125
the Mid-Atlantic Ridge on the African plate (Figure 1). The RGR has been considered as a wide
126
and complex volcanic feature that formed by the interaction of a mantle plume with the mid-
127
ocean ridge., which should be different from volcanic chains that are typically associated with the
128
passage of a tectonic plate over a stationary hotspot, such as the famous Hawaii-Emperor chain
129
(Sharp et al., 2006; Tarduno et al., 2009). Recently, Precambrian-aged granitic rocks were
AC C
EP
121
ACCEPTED MANUSCRIPT 7
130
dredged by the Shinkai 6500 submersible in 2013, suggesting that the RGR is a microcontinent
131
covered by flood basalts (Fioravanti, 2014; Stica et al., 2014). The first studies on the origin of the RGR were performed by Wilson (1963, 1965), Dietz
RI PT
132
& Holden (1970) and Morgan (1971), who developed the hypothesis that the RGR and Walvis
134
Ridge formed from the movements of the South American and African plates over a stationary
135
hotspot located just below the seafloor spreading center in the South Atlantic Ocean. This
136
hypothesis is not accepted today because recent studies have shown that the volcanism in the
137
South Atlantic is closely linked to the break-up of the western portion of the Gondwana continent
138
and the subsequent evolution of the conjugated margins of Brazil and Africa (O'Connor &
139
Duncan, 1990; Ussami et al., 2013). Geological information about this extensive aseismic plateau
140
was generated in the mid-1970s and early 1980s based on seismic reflection and refraction
141
surveys performed by Leyden et al. (1971) and Leyden (1976), results from the early work of the
142
Deep Sea Drilling Project (DSDP) (Baker, 1983) and interpretations of seismic reflection lines
143
along the DSDP drilling data by Gamboa and Rabinowitz (1981, 1984) and Mohriak et al.
144
(2010). In the 2010s, renewed efforts of Brazilian and international institutions culminated in the
145
dredging of igneous and metamorphic Precambrian rocks (Fioravanti, 2014), which supported the
146
assumption that the RGR is a microcontinent covered by Cretaceous basaltic floods (Stica et al.,
147
2014).
M AN U
TE D
EP
AC C
148
SC
133
The RGR is divided into two segments: western (WRGR) and eastern (ERGR) (Figure 2)
149
(Gamboa and Rabinowitz, 1984). The WRGR has an approximately elliptical shape, parallel to
150
the FZs, and separated from the São Paulo Ridge by the N-S oriented Vema Channel. The hook-
151
shaped ERGR is elongated N-S parallel to a section of the Walvis Ridge near Africa and
152
approximately parallel to the mid-ocean ridge axis. The basement of the WRGR is composed of
ACCEPTED MANUSCRIPT 8
Cretaceous (85 Ma) high-Ti tholeiitic basalts (Fodor et al., 1977a). DSDP borehole 516F was the
154
only borehole to reach the volcanic basement rocks of the rise at a depth of 1,271 m (Figure 2).
155
The RGR formed in the seafloor spreading center below sea level during the Santonian-
RI PT
153
Coniacian period and subsequently underwent thermal subsidence followed by pelagic
157
sedimentation, revealed by seismic data obtained by DSDP and University of Texas Institute of
158
Geophysics (UTIG) surveys in the South Atlantic (Baker, 1983; Gamboa and Rabinowitz, 1984;
159
Mohriak et al., 2010; Ussami et al., 2013). As observed in rock samples dredged at point RC16
160
(Figure 2), an alkaline volcanic episode uplifted the WRGR that yielded turbidite and ash layers
161
during the middle Eocene (46 Ma) (Gamboa and Rabinowitz, 1984). This magmatic event gave
162
rise to guyots and seamounts, comprising an alkaline basaltic suite similar to those of oceanic
163
islands (e.g., Tristan da Cunha and Gough) (Fodor et al., 1977a). After the volcanic structures
164
reached the surface to form volcanic islands, the exposed portion of the WRGR was eroded,
165
which resulted in a new sedimentation stage that was characterized by turbidity currents and
166
landslides due to the extension and movement of the underlying crust. Earthquakes associated
167
with the volcanic activity generated turbiditic flows, whose sediments were deposited on the
168
main ERGR platform in close association with the deposits of volcanic ash that were produced
169
during the eruptions. Finally, the entire province underwent thermal subsidence, and a new period
170
of pelagic sedimentation prevailed over the entire segment (Detrick et al., 1977).
M AN U
TE D
EP
AC C
171
SC
156
Knowledge of the ERGR is limited. According to Gamboa and Rabinowitz (1984), this
172
segment is bordered to the north and south by fracture zones, and seismic data indicate up to 800
173
m thick sediment infill in some areas. The age of the basement in the area of DSDP 21 was
174
estimated at 71 Ma by Detrick et al. (1977) and between 75 and 84 Ma by O'Connor & Duncan
175
(1990) based on biostratigraphy data. However, the DSDP21 borehole did not reach the
ACCEPTED MANUSCRIPT 9
basement. The ERGR is separated from the WRGR by a narrow and confined abyssal plain with
177
depths greater than 4,500 m. Numerous seamounts are randomly distributed throughout this plain
178
(Gamboa and Rabinowitz, 1984). Based on the distribution of the FZs and the distance to the
179
current mid-ocean ridge, Gamboa and Rabinowitz (1984) suggested that a part of the ERGR and
180
its conjugate portion in the Walvis Ridge are coeval and by the same tectonic processes. The two
181
segments of the RGR are believed to have different characteristics and origins because the
182
WRGR does not have a conjugate feature on the African plate.
SC
RI PT
176
184
M AN U
183
2.2. Cruzeiro do Sul Rift (CSR) and Jean Charcot Chain
185 186
The RGR is truncated by an NW-SE oriented structure interpreted as an aborted rift that is 10 to 20 km wide, 1,500 km long, and oriented NW-SE (Fig. 2). This linear morphologic
188
depression is clearly marked on bathymetric and gravity maps and is called the Cruzeiro do Sul
189
Lineament (Souza et al., 1993). This feature is associated with a tectonic-magmatic event that
190
affected the RGR and was possibly triggered by the rearrangement of the tectonic plates between
191
the Paleogene and Neogene (Souza et al., 1993; Mohriak et al., 2010). The Cabo Frio High,
192
which divides the two largest oil provinces in Brazil, the Campos and Santos basins, may be the
193
onshore continuation of the trough of the Cruzeiro do Sul Rift (Sharma et al., 1993). According
194
to seismic interpretations performed by Mohriak et al. (2010), extensional structures in the
195
oceanic crust that form semi-grabens are located on the flanks of this rift and are probably
196
associated with increased NW spreading of the structure during the Paleogene and Neogene.
197
Igneous intrusions and volcanic cones also formed in both the oceanic and continental crusts
198
during the Paleogene, which demonstrates the occurrence of pulses of regional scale tectonic and
AC C
EP
TE D
187
ACCEPTED MANUSCRIPT 10
magmatic activity (Souza et al., 1993). In their seismic interpretations, Mohriak et al. (2010)
200
highlighted thick Eocene-Oligocene sedimentation on the valley and flanks of the rift as well as
201
Eocene volcanic layers (Fodor & Thiede, 1977b). This volcanism was indicated in the seismic
202
sections by inclined reflectors related to the basaltic lava flows, which formed at the beginning of
203
seafloor spreading center. They are located at anomalously shallow depths and probably
204
represented a subaerial island during the Santonian (Baker et al., 1981).
SC
205
RI PT
199
The Jean Charcot seamount chain (JCS crosses the western edge of the RGR at a high angle (Figure 2) and remains completely unsampled, so the age and composition of these
207
volcanoes are unknown. A puzzling geological feature observed in the RGR region is associated
208
with the distribution of these seamounts. The southern portion of seamounts curves sharply to an
209
NE-SW orientation, unlike the northern portion, which is oriented NW-SE and follows the trend
210
of the Cruzeiro do Sul Rift. This rotation was not linked to the passage of the South American
211
plate over a hotspot because its main orientation was to the NW, which reflects the absolute
212
movement of the plate during the Cretaceous (O'Connor & Duncan, 1990; Bryant and Cherkis,
213
1995) as is observed in various seamounts along the Brazilian coast. If this chain were new, it
214
should have the same E-W trend as the young seamount chains in the South Atlantic, such as the
215
Martin Vaz chain, which reflects the recent absolute movement of the South American plate
216
(O'Connor & Duncan, 1990).
218
TE D
EP
AC C
217
M AN U
206
3. DATABASE
219 220 221
3.1. Predicted Bathymetry
ACCEPTED MANUSCRIPT 11
222
The bathymetric data used in this study are from the ETOPO1 project and are available from the database of the National Geophysical Data Center of the United States of America
224
(NGDC). The ETOPO1 project data have a spatial resolution of 2 km and are a compilation of
225
topographical data from the Space Shuttle Radar Topography Mission (SRTM) and global
226
topography data from foreign agencies in regions outside of the SRTM data coverage (Amantes
227
& Eakins, 2009). Bathymetric data from oceanic acquisition and satellite altimetry were also
228
used. The data in the NGDC database are already processed and were compiled and referenced
229
relative to sea level. The regular grids of the ETOPO1 project data were generated with the
230
minimum curvature method with 0.0333° (~ 3.7 km) cells.
M AN U
SC
RI PT
223
231 232
3.2. Gravity Dataset
233
The gravity data are from the World Gravity Map 2012 Project (WGM2012) available
TE D
234
from the global database of the International Gravimetric Bureau (BGI). This database contains a
236
set of gravity surveys at a high spatial resolution (~ 2 km). The regular grids are computed from
237
gravity reference models (EGM2008 and DTU10), and an elevation model is used for ground
238
fixes. The gravity data from the Earth Gravitational Model (EGM2008) is a compilation of
239
ground, airborne and satellite data. The DTU10 data is a combination of twelve years of satellite
240
altimetry survey from several missions (ERS-1 GM + GEOSAT, Topex/Poseidon, JASON-1,
241
ER-2, GFO-ERM, ICESAT) and are described by Bonvalot et al. (2012). The WGM2012 is the
242
first dataset of global gravity anomalies and represents a realistic model of these anomalies that
243
considers the contributions of surface masses (atmosphere, continents, oceans, oceanic islands,
244
lakes, ice caps and ice shelves).
AC C
EP
235
ACCEPTED MANUSCRIPT 12
245
In this study, we used free-air and Bouguer gravity anomalies derived from a spherical harmonic approach to achieve precise calculations on a global scale (Balmino et al.., 2012;
247
Bonvalot et al., 2012). Data is preprocessed in the original database and interpolated using the
248
minimum curvature method with 0.0333° (~ 3.7 km) cells.
249
RI PT
246
The matched filter method was applied to separate the Bouguer gravity anomalies related to deep, intermediate and shallow causative sources. The matched filter method is an
251
enhancement technique for potential anomalies that applies an optimized adjustment of bandpass
252
and azimuthal filters to the power spectrum amplitude of a potential field through separation into
253
two or more equivalent layers (Phillips, 2001). This technique obtained large wavelength (> 102
254
km) Bouguer anomalies associated with deep sources (> 23 km deep), intermediate wavelength
255
(between 34 and 102 km) Bouguer anomalies related to sources with depths between 4 and 14
256
km, and medium to short wavelength (< 34 km) Bouguer anomalies associated with shallow
257
sources (< 4 km deep). The regional-scale Bouguer anomalies allow the analysis of the tectonic
258
and isostatic behavior of the RGR in relation to the lithospheric plate that supports its anomalous
259
magmatism. The intermediate gravity anomalies were useful for mapping lateral density
260
variations due to volcanic activity in the RGR and their relationships with the fracture zones,
261
which partially control the magmatic accommodation mechanisms. The intermediate gravity
262
anomalies also provided information about the thickness variations of the sedimentary packages
263
on the plateau and flanks of the rise.
265 266
M AN U
TE D
EP
AC C
264
SC
250
3.3. Magnetic Data
ACCEPTED MANUSCRIPT 13
The magnetic data are provided by the Commission for the Geological Map of the World
268
(CGMW). These magnetic data comprise the Earth Magnetic Anomaly Grid 2 (EMAG2) project,
269
which displays a regular grid with a spatial resolution of 2 arcminutes (~4 km) that contains
270
magnetic anomalies 4 km above sea level. The magnetic anomalies are compiled from satellite
271
data, marine surveys and airborne surveys in various parts of the world (Maus et al., 2009).
The magnetic anomaly maps for the study area were interpolated using the minimum
SC
272
RI PT
267
curvature method with unit cells of 0.0333° (~3.7 km). The total magnetic intensity (TMI) was
274
reduced to the pole (RTP) to centralize the magnetic anomalies on their sources, similar to the
275
pattern of gravity anomalies, which makes the interpretation easier and more reliable (Blakely,
276
1996). The RTP filter parameters are a magnetic inclination of -47.8° and a magnetic declination
277
of -28.8° at the center of the study area. In addition, for easy viewing of the geomagnetic field
278
reversals in the data, the magnetic anomalies were subjected to a simple binary mathematical
279
operation with negative anomalies of -1 nT and positive anomalies of 1 nT. The tilt derivative
280
(TDR) and horizontal gradient amplitude (HGA) enhancement techniques, which are based on
281
the partial derivatives of the magnetic field in the horizontal and vertical directions, were applied
282
to the potential anomalies. These spectral filters facilitate the mapping of magnetic lineaments for
283
the entire frequency spectrum of the magnetic field.
285 286 287
TE D
EP
AC C
284
M AN U
273
4. GEOTECTONIC RECONSTRUCTION OF THE SOUTH ATLANTIC
Boyden et al. (2008) developed the GPlates software to estimate the movement of tectonic
288
plates over geological time based on the relative positions of the plates at different times. The
289
drift of the plates is inferred based on geological, geophysical and paleogeographic data. This
ACCEPTED MANUSCRIPT 14
technique was applied to the bathymetric, gravity and isochrones data from the seafloor to
291
reconstruct the tectonic evolution of the RGR and Walvis Ridge in the context of the formation of
292
the South Atlantic under the influence of the Tristan da Cunha hotspot.
293
RI PT
290
The process for the interactive reconstruction of tectonic plates in this computational platform follows the following steps: a) loading of vector data (point, line, and polygon data) and
295
raster data (continuous information from a particular area that cannot be easily divided into
296
vector data; in this case, the geophysical data); b) data interpolation into a plate tectonic model;
297
and c) reconstruction of these data in a geological time scale. The data comprised a set of
298
polygons that define the extent of each of the georeferenced tectonic plates. The vector and raster
299
data were interpolated into the reconstruction model of the opening of the South Atlantic by
300
Muller et al. (2008). Moulin et al. (2010) and Heine et al. (2013) published new models of the
301
South Atlantic opening. Although these models are more modern and detailed, both
302
reconstructions cover only the rifting and initial seafloor spreading phase (140-84 Ma), while the
303
RGR onset times have been evaluated mainly between 89 and 55 Ma (e.g., Gamboa and
304
Rabinowitz, 1984; Ussami et al., 2012). Once prepared, the combined geometry of the vector
305
data and geophysical data was continuously reconstructed forward or backward in geological
306
time based on the rotation model for each tectonic plate in these intervals. Each rotation consisted
307
of an axis that passed through the center of the Earth and an angle that rigidly rotated the plate on
308
the spherical surface of the globe. The above mentioned steps were performed with the computer
309
graphics tools of the OpenGL programming interface (Williams et al., 2012). The results of the
310
reconstructions were integrated into software linked to georeferenced information systems (GIS).
311
The RGR and Walvis Ridge reconstructions were performed between 100 and 55 Ma at 5
312
AC C
EP
TE D
M AN U
SC
294
My intervals. Five evolutionary stages were defined: 100-90 Ma (I), 90-80 Ma (II), 80-70 Ma
ACCEPTED MANUSCRIPT 15
(III), 70-60 Ma (IV) and 60-55 Ma (V). The current geographical position of the Tristan da
314
Cunha Island was adopted as the location of the hotspot, and its coordinates were held fixed
315
during the reconstruction of both ridges. However, this approach is slightly more complex for the
316
formation of the Walvis Ridge due to the position of the plume (Pessanha, 2011). This aspect was
317
originally questioned by O'Connor & Roex (1992) and Adam et al. (2007) and was discussed in
318
recent studies that refer to a plume between the Tristan da Cunha and Gough islands, which may
319
have been in constant motion in the mantle (O'Connor & Jokat, 2015). However, in this study, the
320
position adopted for the hotspot was compatible with the proposed model for the formation of the
321
RGR. The plume’s position is represented with a 200 km radius based on the dimensions of the
322
zone of influence of the conduit formed by this plume (O'Connor & Roex, 1992; Pessanha,
323
2011).
M AN U
SC
RI PT
313
327 328 329
5.1. RGR Gravity Signatures
EP
326
5. RESULTS
On the Bouguer anomaly map of the South Atlantic (Figure 3A), an increasing positive
AC C
325
TE D
324
330
regional gravity gradient from NW (continent) to SE (ocean) is associated with crustal thinning
331
of the continental region toward the ocean floor, which is typical of passive continental margins
332
(de Castro et al., 1998, 2012; Unternehr et al., 2010; Stica et al., 2014). This regional anomalous
333
pattern is changed by an extensive semicircular gravity low in the RGR (dashed line in the Figure
334
3A), bounded to the south and north by E-W trending fracture zones. To the west, negative
335
Bouguer anomalies extend eastward to the mid-Atlantic ridge. The low Bouguer gravity zone
ACCEPTED MANUSCRIPT 16
336
(290-455 mGal) suggests a region of thickened crust, regardless its continental (microcontinent)
337
or oceanic (mantle plume) nature. Seven gravity domains (Figure 3B) were mapped based on spatial distributions,
RI PT
338
amplitudes and orientations: 1) the GeD1 domain refers to the continental crust of the South
340
American Platform, composed of medium to long wavelength (~ 36.3 km) Bouguer anomalies
341
with a mean amplitude of 34 mGal; 2) the GeD2 domain includes the region with the main rift-
342
phase depocenters of the marginal basins on the South-Southeast Brazilian coast and is also
343
characterized by a zone of stretched and thinned continental crust, hyper-extended eastwards and
344
limited by a belt of exhumed mantle at the transition to oceanic crust (Zalán et al., 2011; Stica et
345
al., 2014). The eastern limit of the GeD2 domain is roughly coincident with the continental-
346
oceanic boundary (COB) interpreted from ultra-deep 2D seismic sections and gravity data by
347
Zalán et al. (2011, their Figure 1). This gravity domain has long wavelength (~ 82.6 km) Bouguer
348
anomalies with a mean amplitude of 220 mGal and is 500 and 2,500 km long in the NW-SE and
349
NE-SW directions, respectively; 3) the GeD3 domain represents a seaward extension of the
350
GeD2 domain and is associated with an area of typical oceanic crust that is thinner than the
351
transitional continental crust to the west. This domain consists of long wavelength (~ 130 km)
352
Bouguer anomalies with a mean amplitude of 415 mGal and a general NE-SW orientation,
353
similar to GeD2, that follows the inflections of the South American Platform; 4) the GeD4
354
domain is associated with the region where the RGR formed (Figures 1 to 3), revealed by an
355
extensive semicircular gravity low with long wavelength (~ 390 km), amplitudes of
356
approximately 280 mGal, an roughly E-W orientation, and lengths of around 420 and 227 km in
357
the E-W and N-S directions, respectively; 5) the GeD5 and GeD6 domains are linked to the
358
Brazil and Argentina oceanic basins, respectively, on the South American plate. The GeD5
AC C
EP
TE D
M AN U
SC
339
ACCEPTED MANUSCRIPT 17
domain is located NE of the RGR and has long wavelength anomalies (~ 175 km) with a mean
360
amplitude of 528 mGal. The GeD6 domain has the highest Bouguer anomaly values (on the order
361
of 536 mGal) with long wavelengths (~ 290 km) southwest of the RGR; and 6) the GeD7 domain
362
comprises a region between the RGR and the Mid-Atlantic Ridge, where the influence of the
363
Tristan da Cunha plume was limited to the formation of several seamount chains. This domain is
364
located southeast of the RGR and contains long wavelength (~ 133.9 km) Bouguer anomalies
365
with mean amplitude of 450 mGal.
SC
RI PT
359
Focusing on the region of the RGR, long and short wavelength gravity lows highlight the
367
WRGR and ERGR segments (dashed black lines in Figures 4A and 4B). The gravity low extends
368
between the RGR segments and the oceanic basins in adjacent areas (hatched area in Figure 4A).
369
In oceanic lithosphere, negative regional Bouguer anomalies normally reflect the effect of
370
isostatic lithospheric compensation that is associated with thickened and less dense crust, whose
371
nature could be an isolated continental crustal block (microcontinent) or volcanic features
372
associated with mantle plumes. In addition, both maps show a roughly N-S-oriented abyssal
373
valley between the two RGR segments (gray polygon in Figure 4B). Its curved shape to the east
374
is coincident with the dextral movement of the E-W trending fracture zones that cross the whole
375
RGR. In this valley, isolated residual negative anomalies with values between -3 and -16 mGal
376
(located in the blue outlines in Figure 4B) are associated with seamounts and guyots that
377
probably are composed by magmatic rocks with lower density than the host oceanic crust. Local
378
gravity lows suggest the presence of areas with significant sedimentary accumulations in the
379
RGR, which formed interlayered with volcanic structures (red outlines in Figure 4B). The
380
formation of the abyssal valley is associated with a major change in the rotation pole and the
AC C
EP
TE D
M AN U
366
ACCEPTED MANUSCRIPT 18
381
rearrangement of the plates at around 84 Ma, which will be discussed in the simplified
382
temporal/spatial reconstruction of the RGR in the next chapter. In the central portion and flanks of the RGR, residual gravity lows (-7 and -3 mGal; white
RI PT
383
polygons in Fig. 5A) reveal the main depocenters of the Maastrichtian to Pleistocene volcano-
385
sedimentary sequences, which overlay the Santonian-Coniacian volcanic basement (ca. 82-96
386
Ma) drilled in the borehole 516F (Gamboa and Rabinowitz, 1984; Mohriak et al., 2010). In
387
addition, two 1,200-km-long NW-SE-oriented negative anomalies cross the WRGR and ERGR.
388
They are associated with the Cruzeiro do Sul Rift (CSR) and the thick sedimentary infill
389
deposited within the trough (white dashed lines in Figure 5A). The uplifted footwall blocks form
390
considerable local relieves across the rift-bounding escarpment and are capped by a narrower
391
sedimentary sequence than the adjacent areas, as was described by Mohriak et al. (2010). This
392
region corresponds to a set of positive residual Bouguer anomalies with a mean value of 10 mGal
393
(pink polygons in Figure 5A). This gravity pattern on the flanks also indicates an aborted rift that
394
was divided into smaller compartments along its length.
M AN U
TE D
The Cruzeiro do Sul Rift (CSR) is also well defined by two extensive negative free-air
EP
395
SC
384
gravity anomalies (pink dashed line in Figure 5B) of -31 and -141 mGal in the WRGR and
397
ERGR, respectively. Extensive free-air gravity lows occur along the CSR. In the ERGR, these
398
free-air gravity lows reach values up to -150 mGal in the deepest part of the rift (~ 5,800 m). In
399
contrast, the rift escarpments are marked by positive free-air anomalies between 59 and 85 mGal
400
(Figures 5B and 6).
401
AC C
396
Extensive E-W gravity lineaments are observed on the northern, southern and eastern
402
boundaries of the RGR (black dashed line in Figure 5B). This anomalous pattern is represented
403
by alternating positive and negative free-air anomalies associated with the FZs and the structural
ACCEPTED MANUSCRIPT 19
highs and lows of the basement. The rise of the mantle material and its rapid cooling caused
405
topographic elevation on the walls of the FZs. These structural highs may also be associated with
406
vertical tectonism of crustal and upper mantle blocks (Bonnati, 1978), and the valleys probably
407
resulted from the stress regime due to transcurrent movements in the active segments of these
408
fractures (Peive, 2006) near the RGR.
Profiles A-A' and B-B' show the correlation between the Bouguer anomalies, free-air
SC
409
RI PT
404
anomalies, and the bathymetry of the RGR (Figure 6). Profile A-A' crosses the WRGR and the
411
northern portion of the ERGR, whereas Profile B-B' crosses only the southern portion of the
412
ERGR. The lowest amplitudes of the Bouguer anomalies suggest that the Moho is deeper beneath
413
the WRGR and ERGR, especially between distances of 500 and 900 km in Profile A-A' and 400
414
and 750 km in Profile B-B’. In both profiles, gravity minima are observed in the CSR region,
415
whose central valley rose in response to the mechanical subsidence that occurred along this
416
structure when extensional forces divided the RGR during the Cenozoic. Short wavelength free-
417
air positive and negative anomalies show a series of seamounts and guyots near the RGR in the
418
NE portion of Profile A-A' and in the SW and NE portions of Profile B-B'. These volcanic
419
features are typical of oceanic crust affected by hot spots or anomalous magmatism along the
420
fracture zones.
TE D
EP
AC C
421
M AN U
410
Finally, a semicircular positive free-air anomaly (on the order of 20 mGal) is located in
422
the southern portion of the WRGR. This may be associated with a possible volcanic massif that is
423
composed of a set of volcanic cones (white circle in Figure 7). These massifs are associated with
424
volcanic eruptions in the ocean and are morphologically distinct from seamounts; they are
425
generally much larger and form domes with gentle slopes, whereas seamounts are typically high
426
with steeper slopes (Sager et al., 2013; Zhang et al., 2015). The randomly distributed volcanic
ACCEPTED MANUSCRIPT 20
cones imply that secondary eruptions may have occurred on this massif. Zhang et al. (2016)
428
described a similar anomalous pattern on the massif in the Shatsky Rise on the eastern margin of
429
Japan.
430 431
5.2. Magnetic Signatures in the RGR region
433
SC
432
RI PT
427
The typical reversal pattern of marine magnetic anomalies is highlighted on the RTP, TDR and HGA maps (Figures 8 and 9), which record the polarity reversals of the Earth's
435
magnetic field while the oceanic crust was being produced at mid-ocean ridges. Linear and
436
alternating positive and negative anomalies are oriented in the north-south direction and crosscut
437
by E-W trending magnetic lineaments related to fracture zones. The long magnetic stripes are
438
sub-parallel to the South American continental margin and roughly follow their inflections. The
439
magnetic anomalies present a significant inflection to the northeast in the region west of the
440
WRGR to the eastern limit of the ERGR, between the fracture zones Meteor and Rio Grande
441
(Figure 8). The magnetic anomaly related to Isochrone C34 (84 Ma) stands out. It is represented
442
by an extensive, well-defined negative magnetic anomaly that crosses the two segments of the
443
RGR (Figure 8). West of Isochrone C34, the magnetic stripes are wider with lower amplitudes,
444
except for the region encompassing the São Paulo Plateau and the WRGR, which present high
445
amplitudes of the analytical signal (Figure 8B). East of Isochrone C34, these lineaments are
446
oriented N-S, secondary NE-SW and NW-SE. This isochrone characterizes the end of the
447
Cretaceous magnetic quiescence period, during which the frequency of the geomagnetic reversals
448
declined steadily to its lowest point (without reversals) at approximately 84 Ma (Cox, 1973).
449
After a magnetically quiet period between 120 and 84 Ma (Malinverno et al., 2012), the
AC C
EP
TE D
M AN U
434
ACCEPTED MANUSCRIPT 21
450
frequency of these reversals increased continuously (Cox, 1973), and the magnetic anomalies
451
were better defined. In addition, a series of E-W-oriented magnetic lineaments that cross the NNE-SSW
RI PT
452
magnetic anomalies was observed (Figures 9 and 10). These lineaments are associated with FZs
454
and their continuations in the rise region, which are not easily observed in the gravity and
455
bathymetric data. The FZs compartmentalize the RGR and are scars produced by transform faults
456
that cross the mid-ocean ridge, deforming the oceanic crust throughout its extent (Adams, 2002).
457
The mapping of these structures, which are hundreds of kilometers long, is essential for
458
understanding the structural framework of the South Atlantic and, in particular, of the RGR,
459
which is transversally crossed by many of these fracture zones. Approximately eleven major FZs
460
are located between 28.7°S and 35.3°S latitude, and some were not identified in earlier studies,
461
such as those at 28.7°S, 29.2°S, 31.1°S, 31.6°S, 32.1°S, 33.7°S and 35.3°S. Additionally, other
462
FZs were better mapped in the present study, such as the Chuí and Cox FZs (Figures 9D and 10).
463
The nomenclature of the new discovered FZs was defined based on the current latitudes of the
464
expressive lineaments of these FZs.
465
EP
TE D
M AN U
SC
453
6. DISCUSSIONS
467
6.1 GEOTECTONIC EVOLUTION OF THE RIO GRAND RISE (100 - 55 Ma)
468 469
AC C
466
A kinematic reconstruction of the temporal and spatial evolution of the RGR and the Mid-
470
Atlantic Ridge segments was performed based on bathymetric, magnetic and free-air and
471
Bouguer gravity data (Figures 11 to 15). In addition, Figure 10 shows a simplified version of the
472
evolution of the RGR in five intervals during this period based on reversal magnetic pattern
ACCEPTED MANUSCRIPT 22
(isochrons described by Muller et al., 2008) and E-W trending magnetic lineaments (fracture
474
zones). Our plate reconstruction is quite similar with the most recent models (e.g., Moulin et al.,
475
2010; Heine et al., 2013; Pérez-Díaz and Eagles, 2014), with the advantages of being the first to
476
integrate bathymetric, gravity (free-air and Bouguer) and magnetic data and to focus on the
477
region of the RGR.
The plate reconstruction maps of South Atlantic evolution at intervals of 5 My support the
SC
478
RI PT
473
hypothesis that the RGR formed due to the excess magmatism from the Tristan da Cunha Plume,
480
which was located between the Rio Grande (North) and 35.3°S (South) fracture zones (FZs) for
481
approximately 45 Ma between late Cretaceous and Paleocene (ca. 95 - 60 Ma). The existence of a
482
microcontinent at the base of the RGR, as suggested by Stica et al. (2014), does not seem to be
483
supported by the present model. This hypothesis imply that the RGR would be located in the
484
African plate at the onset of the seafloor spreading (Barremian; ~127 Ma) and at some point in
485
the South Atlantic growth, the spreading axis jumped eastwards, leaving behind an abandoned
486
mid-oceanic ridge. Pérez-Díaz and Eagles (2014) interpreted the Vema Channel as an abandoned
487
mid-ocean ridge segment in the South American Plate south of the RGR based only on free-air
488
gravity anomalies. However, there is no evidence of the reversal pattern of magnetic anomalies
489
(Figures 8 to 10) favoring this model. Furthermore, these authors consider the RGR a volcanic
490
bulge owing to the action of the Tristan da Cunha plume, which seems inconsistent with a
491
spreading axis supposed surrounded by continental segments.
TE D
EP
AC C
492
M AN U
479
Nunns (1983) depicted the typical distribution of reversal magnetic pattern of marine
493
anomalies in the North Atlantic Ocean. According to this author, in the Norwegian-Greenland
494
Sea, the Jan Mayen microcontinent, with chaotic magnetic pattern, is surrounded by the
495
abandoned Aegir Ridge to the east (magnetic anomalies 24 to 12) and the active Kolbeinsey
ACCEPTED MANUSCRIPT 23
Ridge to the west (magnetic anomalies 7 to 0). Instead, the magnetic anomalies in the South
497
Atlantic plate, comprising the RGR and the Vema Channel, have a roughly north-south trending
498
distribution bended eastwards at the RGR and symmetric to its counterpart in the African plate.
499
In addition, the absence of symmetric magnetic stripes could indicate another scenario in which
500
the RGR represent a continental ribbon and the Vema Channel an aborted rift, regarding to the
501
model of rifting evolution developed by Péron-Pinvidic and Manatschal (2010) for the North
502
Atlantic. In this case, no oceanic crust would have been created between the São Paulo Ridge and
503
the RGR. However, this model contradicts the oceanic nature of the crust, whose boundary with
504
the continental crust was mapped 250 km to the west in ultra-deep seismic sections by Zalán et
505
al. (2011).
M AN U
SC
RI PT
496
Ussami et al. (2012) discussed the possibility of the RGR being a microcontinent
507
regarding to its 30-km crustal thickness. According to them, essential tectono-magmatic
508
conditions, such as re-rifting of a young continental margin and asymmetric sea-floor spreading,
509
are lacking. On the other hand, flexural analysis of free-air gravity data indicates a construction
510
of the RGR related to magmatic episodes. And thus continental crust slivers, such as those
511
recently discovery (Fioravanti, 2014), may have been raised to the surface during the magma
512
emplacement.
EP
AC C
513
TE D
506
During the interval between the 100 and 90 Ma ages (Figures 11 and 12), the center of the
514
hotspot was initially located in the region of the mid-ocean ridge segments between the Rio
515
Grande and Porto Alegre FZs during the Cenomanian to Turonian. The segments north and south
516
of the hotspot migrated northwest, and the segments to the southwest, between the Porto Alegre
517
and Chui FZs, appear to have remained fixed but rotated slightly NNE from their original NE-
518
SW orientation. Starting at 95 Ma, the western portion of the RGR began to form between the
ACCEPTED MANUSCRIPT 24
28.7°S and Porto Alegre FZs and gradually developed southwards to the 31.6°S FZ at
520
approximately 90 Ma. The effect of the formation of the aseismic ridge is illustrated by a
521
semicircular Bouguer gravity anomaly low of 270 mGal in regions of bathymetric high that
522
reaches 400 m at the ends of this rise segment (Fig. 11B).
523
RI PT
519
Between the 90 and 80 Ma ages (Figures 12 and 13), the mid-ocean ridge segments
affected by the hotspot magmatism are distributed between the Rio Grande and 32.1°S FZs. On
525
the South American plate, the entire WRGR developed until around 80 Ma (Campanian). Prior to
526
that, the northern portion of the ERGR began to form between the Rio Grande and 29.2°S FZs
527
starting at ca. 85 Ma (Coniacian-Santonian), which is suggested by the presence of an E-W-
528
oriented gravity anomaly minimum of ca. 370 mGal (A in Figure 12B). At ca. 80 Ma
529
(Campanian), a semicircular valley (gray polygon in Figure 13A) apparently formed from north
530
to south on the ERGR-Walvis plateau on the mid-ocean ridge axis between the 29.2°S and Porto
531
Alegre FZs (yellow dashed line in Figure 13A). As noted by Gente et al. (2003) and Pessanha
532
(2011), this valley represents a plateau rupture related to a period of varying magmatic
533
contributions from the Tristan da Cunha hotspot. The local magmatism decrease differs from
534
what occurs in the Azores, where the increasing distance between the hotspot and ridge axes is
535
the main cause of the decrease in magma supply and of the oceanic plateau break-up. The
536
reconstruction plate model reinforces that during the evolution of the RGR-Walvis system in this
537
period, the main cause of the RGR-Walvis plateau break-up in this area was probably the
538
decrease in the magma supply in these ridge segments. In addition, the decreased magmatic
539
contribution from the plume in the ridge segments between the 29.2°S and 31.1°S FZs is also
540
indicated by the high gravity anomaly (on the order of 460 mGal), which represents a possible
541
area of thinner crust (gray polygon on the Bouguer map in Figure 13B).
AC C
EP
TE D
M AN U
SC
524
ACCEPTED MANUSCRIPT 25
542
After a major change of the rotational pole of the South American and African plates at 84 Ma (Muller et al., 1997) and its consequent spatial rearrangement, the period between 80 and 70
544
Ma (Figures 13 and 14) is represented by a striking instability of the accretion system of the mid-
545
ocean ridge axes in Campanian to Maastrichtian (Pessanha, 2011). Due to the displacements of
546
the South American and African plates during this period, the plume, which was initially
547
positioned over the ridge axis and impacts the region between the Rio Grande and 31.1°S FZs,
548
began to move under the African plate. However, it continued to provide magmatic material to
549
the ERGR between the 28.7°S and Porto Alegre FZs until approximately 75 Ma and extended the
550
magmatism to the 33.7°S FZ at around 70 Ma. The repositioning of the mid-ocean ridge
551
segments between the Porto Alegre and Chui FZs closer to the hotspot becomes evident during
552
this period in the evolutionary reconstruction of the RGR, as shown between intervals II and III
553
of Figure 10. These segments maintained a NNE-SSW orientation but rotated significantly
554
counterclockwise in relation to geographic north, while the segments were rearranged, and new
555
seafloor formed. Finally, the topographic plain formed between the WRGR and ERGR between
556
80 and 70 Ma (Campanian to Maastrichtian) due to the combination of the rearrangement of the
557
lithospheric plates and the migration of ridge segments closer to the hotspot during this period.
558
This reconfiguration occurred in two regions, one between the Rio Grande and Porto Alegre FZs
559
and another to the south between the Porto Alegre and Montevideo FZs. According to Muller et
560
al. (2008), these regions have different spreading rates (≈ 30 and 110 km/My, respectively) and
561
asymmetries (≈ 60% and 90%, respectively). The difference between these rates explains the E-
562
W extent in the southern portion of the plain.
563 564
AC C
EP
TE D
M AN U
SC
RI PT
543
Between 70 and 60 Ma (Figures 14 and 15), the ridge segments between the 31.1°S and Cox FZs were rearranged near the hotspot with a predominant N-S orientation in Maastrichtian to
ACCEPTED MANUSCRIPT 26
Paleocene. During the 65 Ma period, the north and central portions of the ERGR segment had
566
already formed, and the influence of the plume on the ridge segments between Cox and 35.3°S
567
FZs resulted in the development of the southern portion of the ERGR. During the same period,
568
the Chuí and 32.1°S FZs may have stopped developing, or their signatures were not clear in the
569
magnetic data (Figures 8 and 9).
Finally, the southern portion of the ERGR developed until approximately 55 Ma (Figure
SC
570
RI PT
565
15B). During this period (Eocene), the plume was located under the African plate, which reduced
572
its influence on the South American plate side, and it was limited to forming isolated seamounts.
573
However, intense volcanic activity continued in the African plate, which generated the western
574
portions of the Walvis Ridge. The presence of a positive Bouguer gravity anomalies on the order
575
of 470 mGal (gray polygon in Figure 15D) suggests that the oceanic crust in this region of the
576
South American plate has the normal thickness (~ 10 km) reflecting the absence of the magmatic
577
influence of the Tristan da Cunha Plume.
TE D
M AN U
571
578
6.2 TECTONIC SUBDIVISION OF THE CRUZEIRO DO SUL RIFT
EP
579 580
The aborted Cruzeiro do Sul Rift (CSR) is well defined in the gravity data by extensive
AC C
581 582
negative Bouguer and free-air anomalies that follow the NW-SE axis of the central valley
583
(Figures 5 and 6). The flanks of the rift valley progressively decrease towards the southeastern
584
portions of the rift followed by a decrease in the gradient of the free-air anomalies (Figure 7),
585
which indicates that subsidence along this structure progressively increased in this direction. In
586
contrast, the magnetic signature of the rift is not well defined in the dataset used in this study
587
(Figure 8).
ACCEPTED MANUSCRIPT 27
The CSR developed in an extensive deformation zone that cuts the RGR at a high angle as
589
suggested by Mohriak et al. (2010). The aborted rift region represents an area that accommodated
590
the deformation of the extensional faults of the South Atlantic rifting. These movements between
591
large crustal blocks in the accommodation areas of the southeastern margin of the South
592
American Platform played a fundamental role on the magmatic history and development of the
593
Santos and Campos marginal basins, both prior to and after breakup (Mohriak et al., 2010; Quirk
594
et al., 2013).
SC
The breakup and seafloor spreading of the South Atlantic Ocean are a combination of
M AN U
595
RI PT
588
extensional tectonics and dextral shear movement associated with transform faults between the
597
South American and African plate segments that include the RGR (Nurnberg and Muller, 1991;
598
Mohriak et al., 2010). Therefore, the transfer zone in which the CSR developed had an NE-SW
599
oriented extensional component that is oblique to the regional dextral strike-slip movements of E-
600
W trending fracture zones, which affected its development (Figure 16A). These forces along the
601
FZs were possibly activate at the onset of the rift in Barremian-Aptian (Mohriak, 2010) and after
602
during the seafloor fragmented the CSR in large segments between the Paleogene and Neogene
603
ages (Figure 17). According to Cobbold (2001), the entire southeastern margin of South America
604
was reactivated by dextral and sinistral strike-slip movements along the extensional faults of the
605
South Atlantic opening and some segments of transfer faults near the Brazilian coast. These
606
reactivations occurred during the Upper Cretaceous, Eocene and Neogene (Cobbold, 2001) and
607
are correlated with changes in the convergence conditions (direction and speed) of the Andean
608
margin of the South American continent. The tectonic reactivations during the Eocene and
609
Neogene affected the region containing the Cenozoic rift basins along the southeastern margin in
610
Brazil (Zalán and Oliveira, 2005). As well as deeper areas of the South Atlantic are also
AC C
EP
TE D
596
ACCEPTED MANUSCRIPT 28
deformed by these tectonic events, as indicated by the prominent inflections observed along the
612
CSR that developed over the RGR (I and II in Figure 16A). These inflections coincide with areas
613
of FZs and might represent reactivations of old pre-existing transform faults.
614
RI PT
611
The CSR was divided into two segments mainly by the combination of the transcurrent movements between the FZs and the transfer zone in which the rift formed. The northern segment
616
(NCSR) crosses the northwestern portion of the RGR, and the southern segment (SCSR) is
617
located in the southern region of the eastern portion of the rise (Figure 16B). Both segments were
618
rotated counterclockwise in response to the dextral transcurrent forces of the 29.2°S, Cox and
619
Meteor FZs. In the South Atlantic oceanic crust, the NW-SE orientation of the continental shear
620
zones of the Precambrian basement grain is realigned with the E-W orientation of the FZs
621
(Mohriak, 2003), and this alignment is observed in both segments of the rift. Part of the NCSR
622
that formed in the accommodation zone becomes parallel to the Cox FZ at 33.1°S latitude and
623
passes through the eastern portion of the RGR (Fig. 16). In turn, the SCSR segment becomes
624
parallel to the Meteor FZ at 34.5°S latitude.
M AN U
TE D
In addition, the NCSR and SCSR can be subdivided into smaller compartments, which
EP
625
SC
615
were also rotated counterclockwise by strike-slip movements of the FZs. The WRGR and ERGR
627
consist of four and three compartments, respectively (I and II in Figure 17). The compartment
628
between the Porto Alegre and 31.1°S FZs had the greatest rotation along the CRS. In this
629
compartment, a portion of the rift was shifted by a sinistral strike-slip force along a NE-SW-
630
oriented structure (III in Figure 17). This structure is well defined in the TDR map (A in Figure
631
17 III), and it may be related to an extensional fault that formed or was reactivated during the
632
opening of the CRS or an old ridge axis that was reactivated during the development of the rift.
633
AC C
626
ACCEPTED MANUSCRIPT 29
634
6.3 CURVATURE OF THE SOUTHERN PORTION OF THE JEAN CHARCOT CHAIN
635
Similar to the Cruzeiro do Sul Rift (CSR), the Jean Charcot seamount chain (JCC) is not
637
well defined in the magnetic data. However, these structures appear generally well-defined as a
638
series of short-wavelength positive free-air gravity anomalies oriented in NW-SE and NE-SW
639
directions in the northern and southern segments, respectively (Figure 18). They are associated
640
with around 1,000 km long semicircular volcanic structures identified in the bathymetric data
641
(Figure 2). It extends from the Brazilian continental margin for approximately 500 km to the
642
RGR, where it curves sharply SW for another 500 km. Many questions remain about the
643
seamounts on the curved segment, especially about their origin. They are oriented NE-SW, which
644
is opposite to the NW-SE trend of the northern segment of the chain (Figure 2).
M AN U
SC
RI PT
636
One explanation for the differences between the trends of the Jean Charcot Chain
646
segments can be extracted from an analogy with several magmatic events that occurred along the
647
southeastern margin of Brazil. Analyses of seismic sections by Oreiro et al. (2006) indicate that
648
the post-salt magmatic events on the southeastern margin of the South American Platform were
649
the result of the reactivation of deep fault zones, many of which formed during the break-up of
650
the Gondwana supercontinent and by changes in the rotation pole of the South American plate
651
between 84 to 50 Ma. These magmatic events occurred near the intersections between NW-SE-
652
oriented fault zones (strike-slip faults) and NE-SW-oriented fault zones (normal faults) and along
653
the lineament formed by the Poços de Caldas – Cabo Frio alkaline massifs in southeastern Brazil.
654
These data suggest that the magmatism did not occur due to the movement of the plate over a
655
mantle plume because the reactivated faults (mainly the NW-SE-oriented strike-slip faults)
AC C
EP
TE D
645
ACCEPTED MANUSCRIPT 30
656
appear to have cut the entire lithosphere, reached the asthenosphere and caused partial melting of
657
the mantle by pressure relief. A similar situation may have occurred in the southern portion of the Jean Charcot Chain
RI PT
658
but with a peculiar feature. The seamounts in this chain are located with or very close to the FZs,
660
which cross this area, without however presenting a clear alignment whit these E-W trending
661
structures (Figure 18A). On the other hand, the series of seamounts roughly coincides with the
662
magnetic lineaments, especially between Meteor and Rio Grande FZs (Figures 18B and C). The
663
intersections between these reactivated FZs and a likely NE-SW-oriented region of extensional
664
deformation, which developed or was also reactivated during the opening of the CSR, may have
665
formed a large zone of crustal weakness (gray dashed line in Figures 18A and B). In this context,
666
these intersections must have served as conduits for the magmatic rises that formed the southern
667
portion of the mountain range. However, the Jean Charcot Chain and the distribution of its
668
seamounts still require additional detailed studies.
669
671
9. CONCLUSIONS
EP
670
TE D
M AN U
SC
659
The application of enhancement techniques to gravity and magnetic anomalies from
673
global potential models allowed new regional geophysical mapping of the Rio Grande Rise
674
(RGR) in the South Atlantic. The regional Bouguer low indicates a thicker crust related to normal
675
oceanic crust in the adjacent Brazil and Argentina oceanic basins. A crustal root (microcontinent
676
or oceanic origin) beneath the RGR is expected in response to the effect of isostatic compensation
677
of the 2000-m height large elliptical bulge. Furthermore, residual Bouguer lows allow dividing
678
the RGR in two segments (WRGR and ERGR) and indicate regions associated with thick
AC C
672
ACCEPTED MANUSCRIPT 31
sedimentary packages that deposited over the RGR. Finally, the Cruzeiro do Sul Rift (CSR) was
680
well defined by two extensive NW-SE-oriented negative Bouguer and free-air anomalies across
681
the WRGR and ERGR.
682
RI PT
679
The new magnetic maps show the typical reversal magnetic pattern of the oceanic crust, except in the RGR, where the magnetic response is more complex and the N-S orientation of the
684
magnetic anomalies markedly bend to NE-SW. Magnetic anomalies also allowed precisely
685
mapping of FZs throughout the rise region. Approximately eleven FZs observed in the magnetic
686
data compartmentalized the entire rise and the CSR. Based on the processed magnetic anomalies
687
and magnetic isochrones, we identified the spreading pattern of the oceanic crust and its temporal
688
relationship with the formation of the RGR during the Albian to Eocene (~100-55 Ma). In face to
689
the finding of continental rocks recently dredged on the rise, speculation about the RGR being a
690
microcontinent has been renewed. However, the reversal magnetic pattern unravels no evidence
691
of aborted spreading center, indicative of ridge jumps and microcontinent separation. An
692
alternative explanation is that slivers of continental blocks rose during the magma emplacement.
M AN U
TE D
The temporal-spatial evolution of the RGR, Walvis Ridge and Tristan da Cunha Plume
EP
693
SC
683
was analyzed with a plate reconstruction model. The proposed model integrates gravity, magnetic
695
and bathymetric data and covers the interval from 100 to 55 Ma (Albian to Eocene ages), which
696
represents a period of rapid expansion of the seafloor in the South Atlantic. Our reconstruction
697
model is consistent with the modern models of the South Atlantic opening, but it is the only one
698
to specifically focus on the RGR evolution. The proposed model supports the idea that the
699
rearrangement of the segments of the Mid-Atlantic Ridge near the Tristan da Cunha hotspot
700
combined with the additional supply of magma from the hotspot generated the anomalously high
701
and semicircular morphology of the RGR. The extensive FZs may have formed barriers or
AC C
694
ACCEPTED MANUSCRIPT 32
702
conduits for the magmatism from the interaction of the plume with the segments of the mid-ocean
703
ridge during the formation of the RGR. In addition, the FZs, revealed by magnetic lienaments, coincide with the observed
RI PT
704
inflections along the CSR. These inflections might represent pre-existing transform fault zones
706
that were reactivated during the tectonic-magmatic event that formed the aborted rift (Paleogene-
707
Neogene). The division of the rift into two segments (NCSR and SCSR) was based on the
708
significant offsets along the rift. Other smaller counterclockwise-rotated compartments were also
709
observed in the NCSR and SCSR and represent a response to strike-slip dextral movements in the
710
FZs combined with a NE-SW trending extensional regime, which was responsible for the rift
711
formation in the RGR.
M AN U
SC
705
The gravity and magnetic maps also shed light on the origin of the Jean Charcot Chain.
712
The alignment of the series of seamounts, mainly in the southern segment and part of the northern
714
segment to the Rio Grande FZ, with magnetic anomalies suggests that the mechanism of
715
emplacement of these magmatic bodies was influenced by the seafloor spreading of the mid-
716
Atlantic Ridge. This mechanism was still strongly controlled by FZs, which strongly affect this
717
region.
719 720
EP
AC C
718
TE D
713
ACKNOWLEDGMENTS
This research was supported by the IODP/CAPES Program [Project n. 13/2014]. The
721
authors thank the International Gravimetric Bureau (IBG), Commission for the Geological Maps
722
of the World (CCGM) and National Oceanic and Atmospheric Administration (NOAA) for
723
geophysical and bathymetric datasets used in this study. The authors are grateful to the two
ACCEPTED MANUSCRIPT 33
724
anonymous reviewers, whose comments significantly improved the manuscript. ILGG and DLC
725
thank the PRH-PB 229 Scholarship Program (Petrobras) and CNPq for their respective grants.
RI PT
726
REFERENCES
728
Abdelsalam, M. G., Katumwehe, A. B., Atekwana, E. A., Le Pera, A. K., Achang, M. 2016. The
729
Paleoproterozoic Singo granite in the South-Central Uganda revelead as a nested igneous ring
730
complex using geophysical data. Journal of African Earth Sciences, 116, p. 198-212.
SC
727
M AN U
731 732
Adam, C., Vidal, V., Escartin, J. 2007. 80-myr history of buoyancy and volcanic fluxes along the
733
trails of the Walvis and St. Helena hotspots (South Atlantic). Earth and Planetary Science Letters,
734
261(3-4), p. 432-442.
735
Alves, E.C. 1981. Estruturas da margem continental sul brasileira e das áreas oceânicas e
737
continentais, adjacentes. In: PETROBRAS. Estruturas e tectonismo da margem continental
738
brasileira, e suas implicações nos processos sedimentares e na avaliação do potencial de recursos
739
minerais; relatório final. Rio de Janeiro, CENPES/DINTEP. Projeto REMAC 9, p. 145-170.
EP
AC C
740
TE D
736
741
Alves, E. C. 2002. Zonas de fraturas oceânicas e suas relações com a compartimentação tectônica
742
do sudeste do Brasil. Tese de Doutorado. Departamento de Geologia da Universidade Federa do
743
Rio de Janeiro, 247 p.
744 745
Amante, C. and B. W. Eakins. 2009. ETOPO1 1 Arc-Minute Global Relief Model: Procedures,
746
Data Sources and Analysis. NOAA Technical Memorandum NESDIS NGDC-24, March, 19 p.
ACCEPTED MANUSCRIPT 34
747
Asmus, H. E. 1975. Controle estrutural da deposição mesozóica nas bacias da margem
749
continental brasileira. Rev. Bras. Geoci., 5, p. 160–174.
RI PT
748
750
Asmus, H. E., Ferrari, A. L. 1978. Hipótese sobre a causa do tectonismo cenozóico da região
752
Sudeste do Brasil. Rio de Janeiro: PETROBRAS. Série Projeto REMAC 4, p.75–88.
SC
751
753
Asmus, H. E. 1984. Geologia da margem continental brasileira. In: SCHOBBENHAUS, C.;
755
CAMPOS, D.A.; DERZE, G.R.; ASMUS, H.E. coords. Geologia do Brasil. Texto explicativo do
756
mapa geológico do Brasil e da área oceânica adjacente incluindo depósitos minerais. Brasília:
757
DNPM, p. 443–472.
M AN U
754
758
Balmino, G., Vales, N., Bonvalot, S. and Briais, A., 2012. Spherical harmonic modeling to ultra-
760
high degree of Bouguer and isostatic anomalies. Journal of Geodesy. July 2012, Volume 86,
761
Issue 7, p. 499-520.
EP
762
TE D
759
Baker, P. F., Carlson, R. L., Johnson, D. A., Cepek, P., Coulbourn, W., Gamboa, L. A.,
764
Hamilton, N., Melo, U., Pujol, C., Shor, A. N., Suzyumov, A. E., Tjalsma, L. R. C., Walton, W.
765
H. and Weiss, W. 1981. Deep Sea Drilling Project Leg 72: Southwest Atlantic paleocirculation
766
and Rio Grande Rise tectonics. Geol. Soc. Am. Bull., 92: p. 294–309.
767
AC C
763
ACCEPTED MANUSCRIPT 35
Baker, P. F. 1983. Tectonic evolution and subsidence history of the Rio Grande Rise. In: Baker,
769
P. F., Carlson, R. L. et al. (eds) Initial Reports of the Deep Sea Drilling Project, Volume 72. US
770
Government Printing Office, Washington, DC, p. 953–976.
RI PT
768
771
Blakely, R.J., 1996. Potential theory in gravity and magnetic applications. 2nd ed. Cambridge
773
University Press, Cambridge (UK), 441 p.
SC
772
774
Bonatti, E. 1978. Vertical Tectonism in Oceanic Fracture Zones. Earth Planet. Sci. Lett., 37, p.
776
369-379
M AN U
775
777
Bonvalot, S., Balmino, G., Briais, A., Kuhn, M., Peyrefitte, A., Vales, N., Biancale, R., Gabalda,
779
G., Reinquin, F., Sarrailh, M., 2012. World Gravity Map. Commission for the Geological Map of
780
the World. Eds. BGI-CGMW-CNES-IRD, Paris, 8 p.
TE D
778
781
Boyden, J. A., Müller, R. D., Gurnis, M., Torsvik, T. H., Clark, J. A., Turner, M., Ivey-Law, H.,
783
Watson, R. J., and Cannon, J. S. 2011. Next-generation plate-tectonic reconstructions using
784
GPlates. in: Geoinformatics: Cyberinfrastructure for the Solid Earth, edited by: Keller, G. R. and
785
Baru, C., Cambridge University Press, Cambridge, p. 95–114.
AC C
786
EP
782
787
Bryan, P.C., Cherkis, N.Z. 1995. The Bahia Seamounts: test of a hotspot model and a preliminary
788
South American Late Cretaceous to Tertiary Apparent Polar Wander Path. Tectonophysics 241,
789
p. 317–340.
790
ACCEPTED MANUSCRIPT 36
791
Cainelli, C. & Mohriak, W.U. 1999. Some remarks on the evolution of sedimentary basins along
792
the Eastern Brazilian continental margin. Episodes, vol. 22, n. 3, p. 206–216.
RI PT
793
Chang, J-H., Hsieh, H-H., Mirza, A., Chang, S-P., Hsu, H-H., Liu, C-S., Su, C-C., Chiu, S-D.,
795
Ma, Y-F., Chiu, Y-H., Hung, H-T., Lin, Y-C., Chiu, C-H. 2016. Crustal structure north of the Itu
796
Aba Island (Taiping Island), southern margin of the South China Sea. Journal of Asian Earth
797
Sciences, No prelo.
SC
794
M AN U
798 799
Chian, D., Reid, I.D. & Jackson, H.R. 2001. Crustal structure beneath Orphan Basin and
800
implications for nonvolcanic continental rifting. Journal of Geophysical Research, 106 (B6), p.
801
10923–10940.
802
Cobbold, P. R.; Meisling, K. E.; Mount, V. S. 2001. Reactivation of an obliquely rift-ed margin,
804
Campos and Santos basins, southeastern Brazil. Am. Ass. Petr. Geol. Bull., Tulsa, Okla., v. 85, n.
805
11, p. 1925–1944.
EP
806
TE D
803
Cox, A. 1973. Plate tectonics and geomagnetic reversals. San Francisco, California: W. H.
808
Freeman. pp. 138–145, 222–228.
809
AC C
807
810
De Castro, D.L., Medeiros W.E., Jardim de Sá E.F. & Moreira J.A.M. 1998. Mapa gravimétrico
811
do Nordeste Setentrional do Brasil e margem continental adjacente: interpretação com base na
812
hipótese de isostasia. Rev. Bras. de Geof., 16 (2/3), 115–131.
813
ACCEPTED MANUSCRIPT 37
De Castro, D. L., Bezerra, F. H. R., Sousa, M. O. L., and Fuck, R. A., 2012. Influence of
815
Neoproterozoic tectonic fabric on the origin of the Potiguar Basin, northeastern Brazil and its
816
links with West Africa based on Gravity and Magnetic Data, Journal of Geodynamics, 54, 29–42.
RI PT
814
817
Detrick, R. S., Sclater, J. G. and Thiede, J. 1977. The subsidence of aseismic ridges. Earth and
819
Planetary Science Letters, 34, 185–196.
SC
818
820
Dietz, R.S. and Holden, J.C., 1970. Reconstruction of Pangea: breakup and dispersion of the
822
continents, Permian to present. J. Geophys. Res., 75, p.4939–4956.
M AN U
821
823
Estrella, G. O. 1972. O estágio rift nas bacias marginais do Leste Brasileiro. XXVI Congr. Bras.
825
Geol. da S.B.G., Belém-Pará, v.3, p. 29–34.
826
TE D
824
Fairhead, J. D., Marjorie, W. 2005. Plate tectonic processes in the south atlantic ocean: Do we
828
need deep mantle plumes? Special Paper - Geological Society of America, 388, p. 537–553.
829
EP
827
Fioravanti, C. 2014. Ecos da separação - Grandes blocos de rochas com idades e origens
831
diferentes se combinaram ao formar os dois lados do Atlântico Sul. Pesquisa Fapesp, 224, 58-61.
832
AC C
830
833
Fodor, R. V., Husler, J. W. & Kumar, N. 1977a. Petrology of volcanic rocks from an aseismic
834
rise: implications for the origin of the Rio Grande Rise, South Atlantic Ocean. Earth and
835
Planetary Science Letters, 35, p. 225–233.
836
ACCEPTED MANUSCRIPT 38
Fodor, R.V. and Thiede, J. 1977b. Volcanic breccia from DSDP Site 357: Implications for the
838
composition and origin of the Rio Grande Rise. In: P.R. Supko, K. Perch Nielsen et al., Initial
839
Reports of the Deep Sea Drilling Project, v. 39, U.S. Govt. Printing Office, Washington, D.C., p.
840
537–543.
RI PT
837
841
Fodor, R. V., Mckee, E. H., Asmus, H. E. 1984. K-Ar ages and the opening of the South Atlantic
843
Ocean: basaltic rocks from the Brazilian Margins. Marine Geology, 54 (1/2): M1–M8.
SC
842
M AN U
844 845
Fulfaro, V. J., Saad, A. R., Santos, M. V. and Vianna, R. B. 1982. Compartimentação e evolução
846
tectônica da Bacia do Paraná. Rev. Bras. Geoci, v. 12, n. 4, p. 590–611.
847
Gamboa, L. A. P. and Rabinowitz, P. D. 1981. The Rio Grande Fracture Zone in the western
849
South Atlantic and its tectonic implications. Earth Planet. Sci. Lett., 52, p. 410-418
TE D
848
850
Gamboa, L. A. P. and Rabinowitz, P. D. 1984. The evolution of the Rio Grande Rise inthe
852
southwest Atlantic Ocean. Mar. Geol. 58, p. 35–38.
EP
851
AC C
853 854
Gibson, S. A., Thompson, R. N, Day, J. A., Humphris, S. E and Dickins, A. P. 2005. Melt-
855
generation processes associated with the Tristan mantle plume: Constraints on the origin of EM-
856
1. Ear. Planet. Sci. Lett., v. 237, p. 744–767.
857 858
Heine, C., Zoethout, J., Müller, R. D. 2013. Kinematics of the South Atlantic rift. Solid Earth, 4,
859
215–253.
ACCEPTED MANUSCRIPT 39
860
Kaji, A. O., Guerra, J. V., Fernandes, A. M., De Oliveira, R. F., Silva, C. G. and Dos Reis, A. T.
862
2011. Potencial de transporte sedimentar pelas correntes de fundo na região do Canal de Vema
863
(Atlântico Sul). Revista Brasileira de Geofísica, 29(2), p. 385-400.
864
RI PT
861
Kodaira, S., Mjelde, R., Gunnarsson, K., Shiobara, H. & Shimamura, H. 1998. Structure of the
866
Jan Mayen microcontinent and implications for its evolution. Geophysical Journal International,
867
132, p. 383–400.
M AN U
SC
865
868 869
Leyden, R., Ludwig, W., Ewing, M. 1971. Structure of continental margin off Punta del Este,
870
Uruguay, and Rio de Janeiro, Brazil. Assoc. Petr. Geol. Bull., 55:2, p. 161–2.173.
871
Leyden, R. 1976. Salt distribution and crustal models for the eastern Brazilian margin. Anais
873
Acad. Bras. Ciênc., v. 48, p. 159–168.
EP
874
TE D
872
Macedo, J. M. 1989. Evolução tectônica da Bacia de Santos e áreas continentais adjacentes. Bol.
876
Geoc. Petrobrás, Rio de Janeiro, v. 3, no 3, p. 159–173.
877
AC C
875
878
Malinverno, A., Hildebrandt, J., Tominaga, M., Channell, J.E.T., 2012. M-sequence geo-
879
magnetic polarity time scale (MHTC12) that steadies global spreading rates and incorporates
880
astrochronology constraints. J. Geophys. Res.117, B06104.
881
ACCEPTED MANUSCRIPT 40
Maus, S., U. Barckhausen, H. Berkenbosch, N. Bournas, J. Brozena, V. Childers, F. Dostaler, J.
883
D. Fairhead, C. Finn, R. R. B. von Frese, C. Gaina, S. Golynsky, R. Kucks, H. Lühr, P. Milligan,
884
S. Mogren, R. D. Müller, O. Olesen, M. Pilkington, R. Saltus, B. Schreckenberger, E. Thébault,
885
and F. Caratori Tontini, 2009, EMAG2: A 2–arc min resolution Earth Magnetic Anomaly Grid
886
compiled from satellite, airborne, and marine magnetic measurements. Geochemistry,
887
Geophysics, Geosystems, 10, no. 8, 12 p.
SC
RI PT
882
888
Mitzusaki, A. M. P. 1986. Rochas igneo-básicas do Neocomiano da Bacia de Campos:
890
Caracterização e comportamento como reservatório de hidrocarbonetos. Dissertação de Mestrado.
891
Instituto de Geofísica, UFRJ, Rio de Janeiro. 104 p.
M AN U
889
892
894
Morgan, W. J., 1971. Convection plumes in the lower mantle. Nature, 230, p. 42–43.
TE D
893
Mohriak, W. U.; Rosendahl, B. R. 2003. Transform Zones In The South Atlantic Rifted
896
Continental Margins. Geological Society, London, Special Publications, 210, p. 211–228.
897
EP
895
Mohriak , W. U., Nóbrega, M., Odegard, M. E., Gomes, B. S., Dickson, W. G. 2010. Geological
899
and geophysical interpretation of the Rio Grande Rise, south-eastern Brazilian margin;
900
extensional tectonics and rifting of continental and oceanic crusts. Geological Society Publishing
901
House for EAGE (European Assoaciation of Geoscientists & Engineers), London, United
902
Kingdom, Stretching the crust and extending exploration frontiers, p. 231–245.
903
AC C
898
ACCEPTED MANUSCRIPT 41
904
Muller, R. D., Royer, G. S. 1997. A digital age map of the ocean floor. SIO Reference Series,
905
93:30.
RI PT
906
Muller R. D., Sdrolias M., Carmen Gaina Walter R. Roest. 2008. Age, spreading rates, and
908
spreading asymmetry of the world’s ocean crust. Geochemistry Geophysics Geosystems, 9 (4):
909
1–19.
SC
907
910
Nunns, A.G. 1983. Plate tectonic evolution of the Greenland-Scotland Ridge and surrounding
912
regions. In: Bott, M.H.P., Saxov, S., Talwani, M. and Thiede, J. (eds). Structure and
913
Development of the Greenland-Scotland Ridge: New Methods and Concepts. Plenum Press, New
914
York, 11-30.
M AN U
911
915
Nurnberg, D., Muller, R. D. 1991. Tectonic evolution of the South Atlantic from late Jurassic to
917
present. Tectonophysics, 191, 27–53.
EP
918
TE D
916
O’Connor, J. M. & Duncan, R. A. 1990. Evolution of the Walvis-Ridge-Rio Grande Rise hotspot
920
system: implications for African and South American plate motions over plumes. Journal of
921
Geophysical Research, 95 (17), 474-502.
922
AC C
919
923
O’Connor J. M., Le Roex A. P. 1992. South Atlantic hot spot-plume systems: distribution of
924
volcanism in time and space. Earth and Planetary Science Letters, 113 (3), 343–364.
925
ACCEPTED MANUSCRIPT 42
926
O'Connor, J. and Jokat, W. 2015. Tracking the Tristan-Gough mantle plume using discrete chains
927
of intraplate volcanic centers buried in the Walvis Ridge. Geology, 43 (8), p. 715-718.
RI PT
928 929
Ojeda, H. A. O., 1982. Structural framework, stratigraphy, and evolution of Brazilian marginal
930
basins. Am. Ass. Petr. Geol. Bull. v. 66, n. 6, p.732–749.
SC
931
Oreiro, S. G. 2006. Magmatismo e sedimentação em uma área na Plataforma Continental de
933
Cabo Frio, Rio de Janeiro, Brasil, no intervalo Cretáceo Superior – Terciário. Bol. Geoc.
934
Petrobras, Rio de Janeiro, v. 14, n.1, p. 95–112.
M AN U
932
935
Peive, A.A. 1976. Vertical Tectonic Movements of The Crust in Transform Fracture Zones of
937
The Central Atlantic. Geotectonics, 40 (1), 25-36.
938
TE D
936
Pérez-Díaz, L., and Eagles, G. 2014. Constraining South Atlantic growth with seafloor spreading
940
data. Tectonics, 33, 1848–1873.
941
EP
939
Péron-Pinvidic, G., and Manatschal, G. 2010. From microcontinents to extensional allochthons:
943
witnesses of how continents rift and break apart? Petroleum Geoscience, 16, 189–197.
944
AC C
942
945
Pessanha, I. B. M., 2011. Evolution Temporelle et Spatiale du Système d’Interaction entre le
946
point chaud de Tristan da Cunha et la Dorsale de l’Atlantique Sud. G´eophysique [physics.geo-
947
ph]. Université de Bretagne occidentale - Brest, Francais, 262 p.
948
ACCEPTED MANUSCRIPT 43
Phillips, J. D. 2001. Designing matched band pass and azimuthal filters for the separation of
950
potential- field anomalies by source region and source type. ASEG Extended Abstracts 2001:
951
15th Geophysical Conference, p. 1–4.
RI PT
949
952
Pushcharovsky, Y. M. 2013. Microcontinents in the Atlantic Ocean. Geotectonics, 47 (4), 241–
954
250.
SC
953
955
Quirk, D. G., Hertle, M., Jeppesen, J. W., Raven, M., Mohriak, W. U., Kann, D. J., Norgaard, M.,
957
Howe, M. J., Hsu, D., Coffey, B., Mendes, M. P. 2013. Rifting, subsidence and continental
958
break-up above a mantle plume in the central South Atlantic. Geological Society, London,
959
Special Publications, 369 (1), 185-214.
960
M AN U
956
Riccommini, C. 2008. Rapid changes of stress-field in the passive continental margin of
962
southeastern Brazil. 33° Int. Geol. Congress, Abstracts, CD-ROM.
EP
963
TE D
961
Sager, W. W., Zhang J., Korenaga J., Sano T., Koppers A.A.P., Mahoney J., Widdowson M.
965
2013. An immense shield volcano within the Shatsky Rise oceanic plateau, northwest Pacific
966
Ocean. Nat. Geosci., 6, p. 976–981.
967
AC C
964
968
Sharp, W. D., Clague, D. A. 2006. 50-Ma initiation of Hawaiian-Emperor bend records major
969
change in Pacific Plate motion. Science, 313 (5791), p.1281–1284.
970
ACCEPTED MANUSCRIPT 44
Souza, K. G., Fontana, R.L., Mascle, J., Macedo, J.M., Mohriak, W.U. & Hinz, K. 1993. The
972
southern Brazilian margin: an example of a South Atlantic volcanic margin. Third International
973
Congress of the Brazilian Geophysical Society, Rio de Janeiro, RJ, 7–11 November 1993.
974
Sociedade Brasileira de Geologia, Rio de Janeiro, 2, p. 1336–1341.
975
RI PT
971
Spector, A. and Grant, F. 1970. Statistical models for interpreting aeromagnetic data. Geophysics,
977
35, p. 293-302.
SC
976
M AN U
978 979
Stica, J.M., Zalán, P.V. and Ferrari, A.L. 2014. The evolution of rifting on the volcanic margin of
980
the Pelotas Basin and the contextualization of the Paraná-Etendeka LIP in the separation of
981
Gondwana in the South Atlantic: Marine and Petroleum Geology, vol. 50, p. 1-21.
982
Syberg, F.R.J. 1972. A Fourier method for the regional-residual problem of potential fields.
984
Geophysical Prospecting, 20, p. 47-75.
TE D
983
EP
985
Stanton, N. S., 2009. Caracterização Crustal da Margem Sudeste Brasileira através de
987
magnetometria e suas implicações tectônicas. Tese de Doutorado, Universidade do Estado do rio
988
de Janeiro, 170 p.
989
AC C
986
990
Tarduno, J. A., Bunge, H.P., Sleep, N., Hansen, U. 2009. The Bent Hawaiian-Emperor Hotspot
991
Track: Inheriting the Mantle Wind. Science, 324, p. 50–53.
992
ACCEPTED MANUSCRIPT 45
Turner, J. P., Wilson, P.G. 2005. Structure and composition of ocean-continent transition at an
994
obliquely divergent transform margin, Gulf of Guinea, West Africa. Petroleum Geoscience, Vol
995
15 2009, p. 305–311.
RI PT
993
996
Unternehr, P., Péron-Pinvidic, G., Manatschal, G., Sutra, E., 2010. Hyper-extended crust in the
998
South Atlantic: in search of a model. Petroleum Geoscience, 16, 207–215.
SC
997
999
Ussami, N., Chaves C.A.M., Marques L.S., Ernesto M. 2013. Origin of the Rio Grande Rise-
1001
Walvis Ridge reviewed integrating palaeogeographic reconstruction, isotope geochemistry and
1002
flexural modelling. Geological Society, London, Special Publications, 369(1), p. 129–146.
M AN U
1000
1003
Williams, S., Müller R.D., Landgrebe, T. C.W., Whittaker, J.M. 2012. An open-source software
1005
environment for visualizing and refining plate tectonic reconstructions using high resolution
1006
geological and geophysical data sets. In: GSA Today, 22, n. 4/5, p. 4-9.
1007
EP
TE D
1004
Wilson, J. T., 1963. Evidence from islands on the spreading of ocean floors. Nature, v. 197, p.
1009
536-538.
AC C
1008
1010 1011
Wilson, J. T., 1965. Submarine Fracture Zones, aseismic ridges and the Interna-tional Council of
1012
Scientific Unions Line: Proposed Western Margin of the East Pacific Ridge. Nature, 207, p. 907–
1013
911.
1014
ACCEPTED MANUSCRIPT 46
1015
Zalán, P. V. and Oliveira, J. A. B., 2005. Origem e evolução estrutural do Sistema de Riftes
1016
Cenozóicos do Sudeste do Brasil. Bol. Geoci. Petrobras, v.13, n. 2, p.269–300.
RI PT
1017
Zalán, P. V., Severino, M. C. G., Rigoti, C. A., Magnavita, L. P., Oliveira, J. A. B., Vianna, A.
1019
R., 2011. An Entirely New 3D-View of the Crustal and Mantle Structure of a South Atlantic
1020
Passive Margin – Santos, Campos and Espírito Santo Basins, Brazil. AAPG Search and
1021
Discovery Article #30177, AAPG Annual Convention and Exhibition 2011, Houston, Texas,
1022
April 10-13, 12 p. (Expanded Abstract).
M AN U
SC
1018
1023
Zhang J., Sager W.W., Korenaga J. 2015. The Shatsky Rise oceanic plateau structure from two-
1025
dimensional multichannel seismic reflection profiles and implications for oceanic plateau
1026
formation. In: Neal CR, Sager WW, Sano T, Erba E (eds) The origin, evolution, and
1027
environmental impact of oceanic large igneous provinces, Geophysical Society of America
1028
Special paper 511, Boulder, p. 103–126.
TE D
1024
EP
1029
Zhang J., Sager W.W., Durkin W. J. 2016. Morphology of Shatsky Rise oceanic plateau from
1031
high resolution bathymetry. Marine Geophysical Research. Advance online publication, in press.
1032 1033 1034
AC C
1030
FIGURE CAPTIONS
1035
Figure 1 - Rio Grande Rise – Walvis Ridge system of aseismic chains in the southern South
1036
Atlantic. The red circles represent the Tristan da Cunha and Gough hotspots.
1037
ACCEPTED MANUSCRIPT 47
Figure 2 – Rio Grande Rise (RGR) and Cruzeiro do Sul Rift (CSR) on the SE margin of South
1039
America. DSDP boreholes: 516F and 21; Seamount Dredging: RC16; WRGR - Western Segment
1040
of the Rio Grande Rise; ERGR – Eastern Segment of the Rio Grande Rise; JCC - Jean Charcot
1041
Seamounts.
RI PT
1038
1042
Figure 3 - Bouguer anomaly map of the Rio Grande Rise (RGR) showing a gravity low in the
1044
region of the aseismic ridge (A). Main gravity domains (GeD1v-GeD6) observed in the RGR
1045
region and the surrounding area (B).
M AN U
SC
1043
1046
Figure 4 - Maps of regional (A) and intermediate (B) Bouguer anomalies of the Rio Grande Rise
1048
(RGR). The dashed black lines on the maps represent the western and eastern segments of the
1049
rise, and the hatched area in A represents the area of interpreted thickened crust under the RGR.
1050
The gray polygon in B represents the abyssal plain between the RGR segments. The red contours
1051
are associated with seamounts and guyots.
EP
1052
TE D
1047
Figure 5 - Residual Bouguer anomaly map (A) showing the locations of the gravity highs
1054
associated with the basement blocks and the flanks of the Cruzeiro do Sul Rift (CSR) (pink
1055
polygons) and the gravity lows associated with areas of thick sedimentary accumulations (blue
1056
polygons) and the aborted rift (white dashed lines). Free-air anomaly map (B) showing gravity
1057
highs (white polygons) associated with the higher Rio Grande Rise (RGR) regions and the flanks
1058
of the aborted rift. The CSR is represented by an extensive negative free-air anomaly (pink
1059
dashed lines). The black dashed line indicates the deep oceanic region dominated by E-W
1060
oriented fracture zones. Profiles A-A' and B-B' are shown in Figure 6.
AC C
1053
ACCEPTED MANUSCRIPT 48
1061 1062
Figure 6 – Bathymetric and gravity profiles A-A' and B-B' across the Rio Grande Rise (RGR).
RI PT
1063
Figure 7 – Overlapping free-air anomalies in the digital terrain model of the Rio Grande Rise
1065
(RGR). The white circle shows the location of the possible volcanic massif.
1066
SC
1064
Figure 8 – Magnetic anomaly reduced to pole (RTP) (A) and RTP anomalies in binary pattern (B)
1068
maps of the Rio Grande Rise (RGR) (dashed lines).
M AN U
1067
1069
Figure 9 –Tilt derivate (TDR) (A) and horizontal gradient amplitude (HGA) (B) maps with
1071
interpreted magnetic lineaments (C) and oceanic fracture zones (D). The continental-oceanic
1072
crust boundary and Isochrone C34 (84 Ma) are represented by white and red lines, respectively.
1073
TE D
1070
Figure 10 – Five evolutionary stages of the Rio Grande Rise (RGR) and the Mid-Atlantic Ridge
1075
from 100 to 55 Ma. Segments 1 to 3 belongs to the WRGR and segments 4 to 7 to the ERGR.
1076
EP
1074
Figure 11 - Temporal-spatial reconstruction of the absolute movements of the South American
1078
and African plates for the periods of 100 Ma (A) and 95 Ma (B). The Mid-Atlantic Ridge is
1079
represented by the black line, and the white circle represents the zone of influence of the Tristan
1080
da Cunha hotspot (radius of 200 km).
AC C
1077
1081 1082
Figure 12 - Temporal-spatial reconstruction of the absolute movements of the South American
1083
and African plates for the periods of 90 Ma (A) and 85 Ma (B). The Mid-Atlantic Ridge is
ACCEPTED MANUSCRIPT 49
1084
represented by the black line, and the white circle represents the zone of influence of the Tristan
1085
da Cunha hotspot (radius of 200 km).
RI PT
1086
Figure 13 - Temporal-spatial reconstruction of the absolute movements of the South American
1088
and African plates for the periods of 80 Ma (A) and 75 Ma (B). The Mid-Atlantic Ridge is
1089
represented by the black line, and the white circle represents the zone of influence of the Tristan
1090
da Cunha hotspot (radius of 200 km).
SC
1087
M AN U
1091
Figure 14 - Temporal-spatial reconstruction of the absolute movements of the South American
1093
and African plates for the periods of 70 Ma (A) and 65 Ma (B). The Mid-Atlantic Ridge is
1094
represented by the black line, and the white circle represents the zone of influence of the Tristan
1095
da Cunha hotspot (radius of 200 km).
1096
TE D
1092
Figure 15 - Temporal-spatial reconstruction of the absolute movements of the South American
1098
and African plates for the periods of 60 Ma (A) and 55 Ma (B). The Mid-Atlantic Ridge is
1099
represented by the black line, and the white circle represents the zone of influence of the Tristan
1100
da Cunha hotspot (radius of 200 km).
AC C
1101
EP
1097
1102
Figure 16 - Free-air gravity anomaly maps showing the tectonic events that controlled the
1103
formation of the Cruzeiro do Sul Rift (CSR) (A) and its northern (NCSR) and southern (SCSR)
1104
segments (B). Areas I and II show major inflections along the CSR and overlaps with the fracture
1105
zones mapped based on the magnetic lineaments.
1106
ACCEPTED MANUSCRIPT 50
Figure 17 – Free-air anomalies map showing the counterclockwise rotations of the compartments
1108
along the Cruzeiro do Sul Lineament in the western and eastern segments of the Rio Grande Rise
1109
(RGR) (white areas I and II). Area III shows a displacement along the rift associated with a
1110
sinistral transcurrent effort in a NE-SW oriented structure, which is well defined on the TDR map
1111
(Figure 9A). The black lines shown in the enlarged regions represent fracture zones.
RI PT
1107
SC
1112
Figure 18 – Free-air anomaly (A) and TDR (B) maps showing the locations of the Jean Charcot
1114
seamount chain (blue - northern segment; red - southern segment) and the FZs, which
1115
compartmentalize this region (black lines). The southern segment (gray area) is shown in detail in
1116
(C).
AC C
EP
TE D
M AN U
1113
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
S0264-8172(17)30255-6
DOI:
10.1016/j.marpetgeo.2017.06.048
Reference:
JMPG 2982
To appear in:
Marine and Petroleum Geology
Received Date: 11 March 2017 Revised Date:
26 May 2017
Accepted Date: 30 June 2017
Please cite this article as: Guerra Galvão, I.L., Lopes de Castro, D., Contribution of global potential field data to the tectonic reconstruction of the Rio Grande Rise in the South Atlantic, Marine and Petroleum Geology (2017), doi: 10.1016/j.marpetgeo.2017.06.048. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT 1
1
Contribution of Global Potential Field Data to the Tectonic Reconstruction of the Rio
2
Grande Rise in the South Atlantic
RI PT
3
Igor Leonardo Guerra Galvão, David Lopes de Castro
5
Programa de Pós-Graduação em Geodinâmica e Geofísica da Universidade Federal do Rio
6
Grande do Norte, Campus Universitário S/N, 59078-970, Natal, Brasil
7
Corresponding Author: David Lopes de Castro ([email protected])
SC
4
9
M AN U
8
Highlights
• The Rio Grande Rise (South Atlantic) was studied based on potential field data
11
• Regional gravity low indicates crustal thickness beneath the Rio Grande Rise
12
• Magnetic anomalies shed light on the Rio Grande Rise tectonic evolution
13
• Oceanic fracture zones control the development of the rise
14
• Simplified reconstruction model of the Rio Grande Rise is proposed and discussed
AC C
EP
TE D
10
ACCEPTED MANUSCRIPT 2
15
ABSTRACT
16
The application of advanced enhancement techniques for geophysical anomalies to global
RI PT
17
gravity (WGM2012) and magnetic (EMAG2) models sheds light on the complex tectonic
19
evolution of the Rio Grande Rise (RGR) in the southern South Atlantic. Long wavelength
20
Bouguer gravity lows indicate a thicker crust beneath of the ridge, whose nature can be related to
21
a microcontinent or an excess of volcanism within the oceanic realm. Recently dredged
22
continental rocks reinforce the hypothesis of a microcontinent or, at least, slivers of continental
23
crust. However, the reserval magnetic pattern of the processed magnetic anomalies provide no
24
evidence of aborted spreading center similar to the well-studied Jan Mayen microcontinent and
25
the surrounding (inactive) Aegir and (active) Kolbeinsey ridges in the North Atlantic Ocean. The
26
reversal magnetic anomalies show a series N-S trending parallel stripes roughly follow the
27
current South American coastline and segmented by E-W oriented oceanic fracture zones (FZs).
28
The magnetic stripes are bended eastwards at the RGR, showing a more complex magnetic
29
pattern similar to that in the Iceland. The aborted Cruzeiro do Sul Rift (CSR) and the Jean
30
Charcot Chain (JCC) are structures that cross the RGR and contribute to the understanding of the
31
tectonic evolution of the South Atlantic Ocean. NW-SE oriented extensive gravity lows and
32
bathymetric valleys, which mark the CSR, are segmented by E-W trending magnetic lineaments
33
related to FZs. This structural configuration suggests that the extensional event, which formed the
34
rift and the seamounts chain, was replaced by strike-slip movements along the FZs. In addition,
35
we constructed a plate kinematic model for the evolution of the RGR based on bathymetric, free-
36
air and Bouguer gravity and magnetic data. Our model comprises five main stages of the RGR
37
formation and evolution between late Cretaceous and Paleocene (ca. 95 - 60 Ma), separated by
AC C
EP
TE D
M AN U
SC
18
ACCEPTED MANUSCRIPT 3
published seafloor isochrones. The proposed model suggests that the RGR was built at the mid-
39
Atlantic ridge by increased magmatism probably related to the Tristan da Cunha hotspot.
40
Keywords: Potential Field; Aseismic Ridge; Tectonics; South Atlantic
RI PT
38
41 42
1. INTRODUCTION
SC
43
Geodynamic processes of continental break-up and seafloor spreading left behind various
45
large features throughout the world's oceans. These structures detach themselves from the normal
46
oceanic crust by forming large semi-circular or sometimes linear regions such as aseismic ridges
47
or rises. Kumar (1979) lists three possible origins for these features: (1) microcontinents, (2)
48
uplifted oceanic crust, and (3) linear volcanic features. However, this same author points to the
49
possible existence of complex structures, yielded by a combination of the tectonic mechanisms
50
mentioned above. The Rio Grande Rise (RGR) is an example of a large oceanic rise in the South
51
American plate, whose complex origin and tectonic evolution still a matter of debate. The RGR is
52
a mountain range that covers 3,000 km² and rises 3,500 m above the surrounding abyssal plain,
53
located 1,500 km off the Brazilian coast between 27°S and 36°S latitude and 27°W and 40°W
54
longitude (Figure 1). It separates the Brazil (North) and Argentina (South) oceanic basins and
55
represents an extensive barrier to sediment transport between these basins through seafloor
56
currents. The Vema and Hunter channels are the main connections between these oceanic basins
57
(Kaji et al., 2011).
58
AC C
EP
TE D
M AN U
44
The RGR is defined as a semicircular aseismic southern segment of the South Atlantic,
59
and his given name was associated with the absence of earthquakes (e.g., Gamboa and
60
Rabinowitz, 1984). This peculiar morphology, similar to other elevated areas, suggests that these
ACCEPTED MANUSCRIPT 4
regions have different geological histories than the adjacent oceanic crust. These regions
62
sometimes represent isolated fragments of continental crust on the seafloor, such as the Orphan
63
Knoll in the North Atlantic (Chian et al., 2001) and the Jan Mayen microcontinent in northern
64
Iceland (Kodaira et al., 1998). On the other hand, most of these aseismic rises, probably such as
65
the RGR (Gamboa and Rabinowitz, 1984), predominantly consist of basaltic oceanic rocks where
66
sampled.
SC
67
RI PT
61
The RGR formed within a series of structures related to the rifting and opening of the South Atlantic Ocean in the early Cretaceous (130 Ma) (Cobbold et al., 2001; Zálan and Oliveira,
69
2005; Ricommini, 2008). The transit of the lithospheric plate over the Tristan da Cunha Plume
70
probably caused the extensional events that resulted in the break-up of the Gondwana
71
supercontinent and resulted on extensive basaltic lava flows during the Neocomian, collectively
72
known as the Paraná-Etendeka large Igneous Province (Fodor et al., 1984; Mitzusaki, 1986;
73
O'Connor and Duncan, 1990). This plume was responsible for generating a thermal event that
74
reached the Western Gondwana continental lithosphere, particularly in southeastern Brazil, where
75
a wide area was uplifted and underwent changes in its rheology to become more ductile and less
76
resistant to stretching (Estrella, 1972; Asmus and Ferrari, 1978; Asmus, 1975, 1984; Ojeda, 1982;
77
Fulfaro et al., 1982; Macedo, 1989).
TE D
EP
AC C
78
M AN U
68
O'Connor and Duncan (1990) associated the formation of the RGR-Walvis Ridge system
79
to the presence of the Tristan da Cunha-Gough Plume, which is located on the spreading axis of
80
the Mid-Atlantic ridge with its large supply of magma. Gibson et al. (2005) noted that the mantle
81
plume is currently located 550 km from the central axis of the range between the Tristan da
82
Cunha and Gough islands. Recent work by Stica et al. (2014, their Fig. 14) considers the Rio
83
Grande Rise to be a continental block covered by thick volcanic deposits of unknown origin
ACCEPTED MANUSCRIPT 5
based on newly dredged rock samples of Precambrian granitic and metamorphic rocks in
85
oceanographic expeditions lead by the Brazilian Geological Survey (BSG) and a pool of
86
Brazilian and international universities in 2011 and 2013 (Fioravanti, 2014). Pushcharovsky
87
(2013) also pointed out that RGR is a microcontinent spatially related to the São Paulo
88
continental prominence and separated by the N-S oriented deep trough of the Vema Channel
89
(Figures 1 and 2).The presence of typically continental crust rocks on seabed, although literally
90
essential, is nevertheless insufficient to define the RGR as a continental fragment of the South
91
American plate. Other key evidences, such as extinct spreading ridge, and jumped spreading
92
center, are still necessary to confirm this new approach. As well, geophysical, geological and
93
geochronological datasets must be reinterpreted and plate tectonic reconstruction models must be
94
reviewed to accommodate an ancient microcontinent 1300 km far away of the current coast line.
95
However, the characteristics of the RGR make it difficult to precisely determine its genesis,
96
especially regarding its depth and the difficulty of processes occurring at such depths. Its
97
geotectonic evolution is still disputed, and information about its structural framework is scarce
98
and requires better explanations.
SC
M AN U
TE D
EP
100
In this study, we apply advanced enhancement techniques for gravity and magnetic
AC C
99
RI PT
84
101
anomalies to investigate the formation of the RGR from the interaction between the tectonic
102
processes of the evolution of both the Mid-Atlantic Ridge and the Tristan da Cunha hotspot
103
during the opening of the South Atlantic in a regional context. The structural control of the
104
oceanic fracture zones (FZs) and the extensional tectonic events associated with the evolution of
105
the extensive failed Cruzeiro do Sul Rift and the formation of the southern portion of the Jean
106
Charcot seamount chain are also examined. We show a reconstruction model of the RGR from
ACCEPTED MANUSCRIPT 6
100 Ma to 55 Ma based on bathymetric, gravity and magnetic data and discuss the origin of this
108
large aseismic rise concerning to crustal thickening, seafloor spreading and structural control by
109
fracture zones. In summary, the novelties of this paper are: 1) crustal thickness beneath the RGR
110
derived from regional Bouguer anomaly; 2) areal distribution of depocenters and seamounts in
111
the aseismic bulge region unraveled by residual Bouguer and free-air anomalies; 3) detailed
112
seafloor spreading and fracture zones mapped using enhanced magnetic anomalies; and 4)
113
detailed plate motion reconstruction of the RGR based on bathymetric, gravity and magnetic data
114
including discussions about the crustal nature of the rise, tectonics of the Cruzeiro do Sul rift and
115
the role of the fracture zones in the context of the South Atlantic opening.
M AN U
SC
RI PT
107
116 117
2. REGIONAL GEOLOGICAL-STRUCTURAL SETTING
118
120
2.1. Rio Grande Rise (RGR)
TE D
119
The RGR is located on the SE coast of South America (Figure 1) and significantly
122
modifies the planar morphology of the abyssal oceanic domain between the Brazil and Argentina
123
oceanic basins in the western South Atlantic (Adams, 1981). The Walvis Ridge, other important
124
tectonic element in the South Atlantic Ocean, is a NE-SW chain located on the opposite side of
125
the Mid-Atlantic Ridge on the African plate (Figure 1). The RGR has been considered as a wide
126
and complex volcanic feature that formed by the interaction of a mantle plume with the mid-
127
ocean ridge., which should be different from volcanic chains that are typically associated with the
128
passage of a tectonic plate over a stationary hotspot, such as the famous Hawaii-Emperor chain
129
(Sharp et al., 2006; Tarduno et al., 2009). Recently, Precambrian-aged granitic rocks were
AC C
EP
121
ACCEPTED MANUSCRIPT 7
130
dredged by the Shinkai 6500 submersible in 2013, suggesting that the RGR is a microcontinent
131
covered by flood basalts (Fioravanti, 2014; Stica et al., 2014). The first studies on the origin of the RGR were performed by Wilson (1963, 1965), Dietz
RI PT
132
& Holden (1970) and Morgan (1971), who developed the hypothesis that the RGR and Walvis
134
Ridge formed from the movements of the South American and African plates over a stationary
135
hotspot located just below the seafloor spreading center in the South Atlantic Ocean. This
136
hypothesis is not accepted today because recent studies have shown that the volcanism in the
137
South Atlantic is closely linked to the break-up of the western portion of the Gondwana continent
138
and the subsequent evolution of the conjugated margins of Brazil and Africa (O'Connor &
139
Duncan, 1990; Ussami et al., 2013). Geological information about this extensive aseismic plateau
140
was generated in the mid-1970s and early 1980s based on seismic reflection and refraction
141
surveys performed by Leyden et al. (1971) and Leyden (1976), results from the early work of the
142
Deep Sea Drilling Project (DSDP) (Baker, 1983) and interpretations of seismic reflection lines
143
along the DSDP drilling data by Gamboa and Rabinowitz (1981, 1984) and Mohriak et al.
144
(2010). In the 2010s, renewed efforts of Brazilian and international institutions culminated in the
145
dredging of igneous and metamorphic Precambrian rocks (Fioravanti, 2014), which supported the
146
assumption that the RGR is a microcontinent covered by Cretaceous basaltic floods (Stica et al.,
147
2014).
M AN U
TE D
EP
AC C
148
SC
133
The RGR is divided into two segments: western (WRGR) and eastern (ERGR) (Figure 2)
149
(Gamboa and Rabinowitz, 1984). The WRGR has an approximately elliptical shape, parallel to
150
the FZs, and separated from the São Paulo Ridge by the N-S oriented Vema Channel. The hook-
151
shaped ERGR is elongated N-S parallel to a section of the Walvis Ridge near Africa and
152
approximately parallel to the mid-ocean ridge axis. The basement of the WRGR is composed of
ACCEPTED MANUSCRIPT 8
Cretaceous (85 Ma) high-Ti tholeiitic basalts (Fodor et al., 1977a). DSDP borehole 516F was the
154
only borehole to reach the volcanic basement rocks of the rise at a depth of 1,271 m (Figure 2).
155
The RGR formed in the seafloor spreading center below sea level during the Santonian-
RI PT
153
Coniacian period and subsequently underwent thermal subsidence followed by pelagic
157
sedimentation, revealed by seismic data obtained by DSDP and University of Texas Institute of
158
Geophysics (UTIG) surveys in the South Atlantic (Baker, 1983; Gamboa and Rabinowitz, 1984;
159
Mohriak et al., 2010; Ussami et al., 2013). As observed in rock samples dredged at point RC16
160
(Figure 2), an alkaline volcanic episode uplifted the WRGR that yielded turbidite and ash layers
161
during the middle Eocene (46 Ma) (Gamboa and Rabinowitz, 1984). This magmatic event gave
162
rise to guyots and seamounts, comprising an alkaline basaltic suite similar to those of oceanic
163
islands (e.g., Tristan da Cunha and Gough) (Fodor et al., 1977a). After the volcanic structures
164
reached the surface to form volcanic islands, the exposed portion of the WRGR was eroded,
165
which resulted in a new sedimentation stage that was characterized by turbidity currents and
166
landslides due to the extension and movement of the underlying crust. Earthquakes associated
167
with the volcanic activity generated turbiditic flows, whose sediments were deposited on the
168
main ERGR platform in close association with the deposits of volcanic ash that were produced
169
during the eruptions. Finally, the entire province underwent thermal subsidence, and a new period
170
of pelagic sedimentation prevailed over the entire segment (Detrick et al., 1977).
M AN U
TE D
EP
AC C
171
SC
156
Knowledge of the ERGR is limited. According to Gamboa and Rabinowitz (1984), this
172
segment is bordered to the north and south by fracture zones, and seismic data indicate up to 800
173
m thick sediment infill in some areas. The age of the basement in the area of DSDP 21 was
174
estimated at 71 Ma by Detrick et al. (1977) and between 75 and 84 Ma by O'Connor & Duncan
175
(1990) based on biostratigraphy data. However, the DSDP21 borehole did not reach the
ACCEPTED MANUSCRIPT 9
basement. The ERGR is separated from the WRGR by a narrow and confined abyssal plain with
177
depths greater than 4,500 m. Numerous seamounts are randomly distributed throughout this plain
178
(Gamboa and Rabinowitz, 1984). Based on the distribution of the FZs and the distance to the
179
current mid-ocean ridge, Gamboa and Rabinowitz (1984) suggested that a part of the ERGR and
180
its conjugate portion in the Walvis Ridge are coeval and by the same tectonic processes. The two
181
segments of the RGR are believed to have different characteristics and origins because the
182
WRGR does not have a conjugate feature on the African plate.
SC
RI PT
176
184
M AN U
183
2.2. Cruzeiro do Sul Rift (CSR) and Jean Charcot Chain
185 186
The RGR is truncated by an NW-SE oriented structure interpreted as an aborted rift that is 10 to 20 km wide, 1,500 km long, and oriented NW-SE (Fig. 2). This linear morphologic
188
depression is clearly marked on bathymetric and gravity maps and is called the Cruzeiro do Sul
189
Lineament (Souza et al., 1993). This feature is associated with a tectonic-magmatic event that
190
affected the RGR and was possibly triggered by the rearrangement of the tectonic plates between
191
the Paleogene and Neogene (Souza et al., 1993; Mohriak et al., 2010). The Cabo Frio High,
192
which divides the two largest oil provinces in Brazil, the Campos and Santos basins, may be the
193
onshore continuation of the trough of the Cruzeiro do Sul Rift (Sharma et al., 1993). According
194
to seismic interpretations performed by Mohriak et al. (2010), extensional structures in the
195
oceanic crust that form semi-grabens are located on the flanks of this rift and are probably
196
associated with increased NW spreading of the structure during the Paleogene and Neogene.
197
Igneous intrusions and volcanic cones also formed in both the oceanic and continental crusts
198
during the Paleogene, which demonstrates the occurrence of pulses of regional scale tectonic and
AC C
EP
TE D
187
ACCEPTED MANUSCRIPT 10
magmatic activity (Souza et al., 1993). In their seismic interpretations, Mohriak et al. (2010)
200
highlighted thick Eocene-Oligocene sedimentation on the valley and flanks of the rift as well as
201
Eocene volcanic layers (Fodor & Thiede, 1977b). This volcanism was indicated in the seismic
202
sections by inclined reflectors related to the basaltic lava flows, which formed at the beginning of
203
seafloor spreading center. They are located at anomalously shallow depths and probably
204
represented a subaerial island during the Santonian (Baker et al., 1981).
SC
205
RI PT
199
The Jean Charcot seamount chain (JCS crosses the western edge of the RGR at a high angle (Figure 2) and remains completely unsampled, so the age and composition of these
207
volcanoes are unknown. A puzzling geological feature observed in the RGR region is associated
208
with the distribution of these seamounts. The southern portion of seamounts curves sharply to an
209
NE-SW orientation, unlike the northern portion, which is oriented NW-SE and follows the trend
210
of the Cruzeiro do Sul Rift. This rotation was not linked to the passage of the South American
211
plate over a hotspot because its main orientation was to the NW, which reflects the absolute
212
movement of the plate during the Cretaceous (O'Connor & Duncan, 1990; Bryant and Cherkis,
213
1995) as is observed in various seamounts along the Brazilian coast. If this chain were new, it
214
should have the same E-W trend as the young seamount chains in the South Atlantic, such as the
215
Martin Vaz chain, which reflects the recent absolute movement of the South American plate
216
(O'Connor & Duncan, 1990).
218
TE D
EP
AC C
217
M AN U
206
3. DATABASE
219 220 221
3.1. Predicted Bathymetry
ACCEPTED MANUSCRIPT 11
222
The bathymetric data used in this study are from the ETOPO1 project and are available from the database of the National Geophysical Data Center of the United States of America
224
(NGDC). The ETOPO1 project data have a spatial resolution of 2 km and are a compilation of
225
topographical data from the Space Shuttle Radar Topography Mission (SRTM) and global
226
topography data from foreign agencies in regions outside of the SRTM data coverage (Amantes
227
& Eakins, 2009). Bathymetric data from oceanic acquisition and satellite altimetry were also
228
used. The data in the NGDC database are already processed and were compiled and referenced
229
relative to sea level. The regular grids of the ETOPO1 project data were generated with the
230
minimum curvature method with 0.0333° (~ 3.7 km) cells.
M AN U
SC
RI PT
223
231 232
3.2. Gravity Dataset
233
The gravity data are from the World Gravity Map 2012 Project (WGM2012) available
TE D
234
from the global database of the International Gravimetric Bureau (BGI). This database contains a
236
set of gravity surveys at a high spatial resolution (~ 2 km). The regular grids are computed from
237
gravity reference models (EGM2008 and DTU10), and an elevation model is used for ground
238
fixes. The gravity data from the Earth Gravitational Model (EGM2008) is a compilation of
239
ground, airborne and satellite data. The DTU10 data is a combination of twelve years of satellite
240
altimetry survey from several missions (ERS-1 GM + GEOSAT, Topex/Poseidon, JASON-1,
241
ER-2, GFO-ERM, ICESAT) and are described by Bonvalot et al. (2012). The WGM2012 is the
242
first dataset of global gravity anomalies and represents a realistic model of these anomalies that
243
considers the contributions of surface masses (atmosphere, continents, oceans, oceanic islands,
244
lakes, ice caps and ice shelves).
AC C
EP
235
ACCEPTED MANUSCRIPT 12
245
In this study, we used free-air and Bouguer gravity anomalies derived from a spherical harmonic approach to achieve precise calculations on a global scale (Balmino et al.., 2012;
247
Bonvalot et al., 2012). Data is preprocessed in the original database and interpolated using the
248
minimum curvature method with 0.0333° (~ 3.7 km) cells.
249
RI PT
246
The matched filter method was applied to separate the Bouguer gravity anomalies related to deep, intermediate and shallow causative sources. The matched filter method is an
251
enhancement technique for potential anomalies that applies an optimized adjustment of bandpass
252
and azimuthal filters to the power spectrum amplitude of a potential field through separation into
253
two or more equivalent layers (Phillips, 2001). This technique obtained large wavelength (> 102
254
km) Bouguer anomalies associated with deep sources (> 23 km deep), intermediate wavelength
255
(between 34 and 102 km) Bouguer anomalies related to sources with depths between 4 and 14
256
km, and medium to short wavelength (< 34 km) Bouguer anomalies associated with shallow
257
sources (< 4 km deep). The regional-scale Bouguer anomalies allow the analysis of the tectonic
258
and isostatic behavior of the RGR in relation to the lithospheric plate that supports its anomalous
259
magmatism. The intermediate gravity anomalies were useful for mapping lateral density
260
variations due to volcanic activity in the RGR and their relationships with the fracture zones,
261
which partially control the magmatic accommodation mechanisms. The intermediate gravity
262
anomalies also provided information about the thickness variations of the sedimentary packages
263
on the plateau and flanks of the rise.
265 266
M AN U
TE D
EP
AC C
264
SC
250
3.3. Magnetic Data
ACCEPTED MANUSCRIPT 13
The magnetic data are provided by the Commission for the Geological Map of the World
268
(CGMW). These magnetic data comprise the Earth Magnetic Anomaly Grid 2 (EMAG2) project,
269
which displays a regular grid with a spatial resolution of 2 arcminutes (~4 km) that contains
270
magnetic anomalies 4 km above sea level. The magnetic anomalies are compiled from satellite
271
data, marine surveys and airborne surveys in various parts of the world (Maus et al., 2009).
The magnetic anomaly maps for the study area were interpolated using the minimum
SC
272
RI PT
267
curvature method with unit cells of 0.0333° (~3.7 km). The total magnetic intensity (TMI) was
274
reduced to the pole (RTP) to centralize the magnetic anomalies on their sources, similar to the
275
pattern of gravity anomalies, which makes the interpretation easier and more reliable (Blakely,
276
1996). The RTP filter parameters are a magnetic inclination of -47.8° and a magnetic declination
277
of -28.8° at the center of the study area. In addition, for easy viewing of the geomagnetic field
278
reversals in the data, the magnetic anomalies were subjected to a simple binary mathematical
279
operation with negative anomalies of -1 nT and positive anomalies of 1 nT. The tilt derivative
280
(TDR) and horizontal gradient amplitude (HGA) enhancement techniques, which are based on
281
the partial derivatives of the magnetic field in the horizontal and vertical directions, were applied
282
to the potential anomalies. These spectral filters facilitate the mapping of magnetic lineaments for
283
the entire frequency spectrum of the magnetic field.
285 286 287
TE D
EP
AC C
284
M AN U
273
4. GEOTECTONIC RECONSTRUCTION OF THE SOUTH ATLANTIC
Boyden et al. (2008) developed the GPlates software to estimate the movement of tectonic
288
plates over geological time based on the relative positions of the plates at different times. The
289
drift of the plates is inferred based on geological, geophysical and paleogeographic data. This
ACCEPTED MANUSCRIPT 14
technique was applied to the bathymetric, gravity and isochrones data from the seafloor to
291
reconstruct the tectonic evolution of the RGR and Walvis Ridge in the context of the formation of
292
the South Atlantic under the influence of the Tristan da Cunha hotspot.
293
RI PT
290
The process for the interactive reconstruction of tectonic plates in this computational platform follows the following steps: a) loading of vector data (point, line, and polygon data) and
295
raster data (continuous information from a particular area that cannot be easily divided into
296
vector data; in this case, the geophysical data); b) data interpolation into a plate tectonic model;
297
and c) reconstruction of these data in a geological time scale. The data comprised a set of
298
polygons that define the extent of each of the georeferenced tectonic plates. The vector and raster
299
data were interpolated into the reconstruction model of the opening of the South Atlantic by
300
Muller et al. (2008). Moulin et al. (2010) and Heine et al. (2013) published new models of the
301
South Atlantic opening. Although these models are more modern and detailed, both
302
reconstructions cover only the rifting and initial seafloor spreading phase (140-84 Ma), while the
303
RGR onset times have been evaluated mainly between 89 and 55 Ma (e.g., Gamboa and
304
Rabinowitz, 1984; Ussami et al., 2012). Once prepared, the combined geometry of the vector
305
data and geophysical data was continuously reconstructed forward or backward in geological
306
time based on the rotation model for each tectonic plate in these intervals. Each rotation consisted
307
of an axis that passed through the center of the Earth and an angle that rigidly rotated the plate on
308
the spherical surface of the globe. The above mentioned steps were performed with the computer
309
graphics tools of the OpenGL programming interface (Williams et al., 2012). The results of the
310
reconstructions were integrated into software linked to georeferenced information systems (GIS).
311
The RGR and Walvis Ridge reconstructions were performed between 100 and 55 Ma at 5
312
AC C
EP
TE D
M AN U
SC
294
My intervals. Five evolutionary stages were defined: 100-90 Ma (I), 90-80 Ma (II), 80-70 Ma
ACCEPTED MANUSCRIPT 15
(III), 70-60 Ma (IV) and 60-55 Ma (V). The current geographical position of the Tristan da
314
Cunha Island was adopted as the location of the hotspot, and its coordinates were held fixed
315
during the reconstruction of both ridges. However, this approach is slightly more complex for the
316
formation of the Walvis Ridge due to the position of the plume (Pessanha, 2011). This aspect was
317
originally questioned by O'Connor & Roex (1992) and Adam et al. (2007) and was discussed in
318
recent studies that refer to a plume between the Tristan da Cunha and Gough islands, which may
319
have been in constant motion in the mantle (O'Connor & Jokat, 2015). However, in this study, the
320
position adopted for the hotspot was compatible with the proposed model for the formation of the
321
RGR. The plume’s position is represented with a 200 km radius based on the dimensions of the
322
zone of influence of the conduit formed by this plume (O'Connor & Roex, 1992; Pessanha,
323
2011).
M AN U
SC
RI PT
313
327 328 329
5.1. RGR Gravity Signatures
EP
326
5. RESULTS
On the Bouguer anomaly map of the South Atlantic (Figure 3A), an increasing positive
AC C
325
TE D
324
330
regional gravity gradient from NW (continent) to SE (ocean) is associated with crustal thinning
331
of the continental region toward the ocean floor, which is typical of passive continental margins
332
(de Castro et al., 1998, 2012; Unternehr et al., 2010; Stica et al., 2014). This regional anomalous
333
pattern is changed by an extensive semicircular gravity low in the RGR (dashed line in the Figure
334
3A), bounded to the south and north by E-W trending fracture zones. To the west, negative
335
Bouguer anomalies extend eastward to the mid-Atlantic ridge. The low Bouguer gravity zone
ACCEPTED MANUSCRIPT 16
336
(290-455 mGal) suggests a region of thickened crust, regardless its continental (microcontinent)
337
or oceanic (mantle plume) nature. Seven gravity domains (Figure 3B) were mapped based on spatial distributions,
RI PT
338
amplitudes and orientations: 1) the GeD1 domain refers to the continental crust of the South
340
American Platform, composed of medium to long wavelength (~ 36.3 km) Bouguer anomalies
341
with a mean amplitude of 34 mGal; 2) the GeD2 domain includes the region with the main rift-
342
phase depocenters of the marginal basins on the South-Southeast Brazilian coast and is also
343
characterized by a zone of stretched and thinned continental crust, hyper-extended eastwards and
344
limited by a belt of exhumed mantle at the transition to oceanic crust (Zalán et al., 2011; Stica et
345
al., 2014). The eastern limit of the GeD2 domain is roughly coincident with the continental-
346
oceanic boundary (COB) interpreted from ultra-deep 2D seismic sections and gravity data by
347
Zalán et al. (2011, their Figure 1). This gravity domain has long wavelength (~ 82.6 km) Bouguer
348
anomalies with a mean amplitude of 220 mGal and is 500 and 2,500 km long in the NW-SE and
349
NE-SW directions, respectively; 3) the GeD3 domain represents a seaward extension of the
350
GeD2 domain and is associated with an area of typical oceanic crust that is thinner than the
351
transitional continental crust to the west. This domain consists of long wavelength (~ 130 km)
352
Bouguer anomalies with a mean amplitude of 415 mGal and a general NE-SW orientation,
353
similar to GeD2, that follows the inflections of the South American Platform; 4) the GeD4
354
domain is associated with the region where the RGR formed (Figures 1 to 3), revealed by an
355
extensive semicircular gravity low with long wavelength (~ 390 km), amplitudes of
356
approximately 280 mGal, an roughly E-W orientation, and lengths of around 420 and 227 km in
357
the E-W and N-S directions, respectively; 5) the GeD5 and GeD6 domains are linked to the
358
Brazil and Argentina oceanic basins, respectively, on the South American plate. The GeD5
AC C
EP
TE D
M AN U
SC
339
ACCEPTED MANUSCRIPT 17
domain is located NE of the RGR and has long wavelength anomalies (~ 175 km) with a mean
360
amplitude of 528 mGal. The GeD6 domain has the highest Bouguer anomaly values (on the order
361
of 536 mGal) with long wavelengths (~ 290 km) southwest of the RGR; and 6) the GeD7 domain
362
comprises a region between the RGR and the Mid-Atlantic Ridge, where the influence of the
363
Tristan da Cunha plume was limited to the formation of several seamount chains. This domain is
364
located southeast of the RGR and contains long wavelength (~ 133.9 km) Bouguer anomalies
365
with mean amplitude of 450 mGal.
SC
RI PT
359
Focusing on the region of the RGR, long and short wavelength gravity lows highlight the
367
WRGR and ERGR segments (dashed black lines in Figures 4A and 4B). The gravity low extends
368
between the RGR segments and the oceanic basins in adjacent areas (hatched area in Figure 4A).
369
In oceanic lithosphere, negative regional Bouguer anomalies normally reflect the effect of
370
isostatic lithospheric compensation that is associated with thickened and less dense crust, whose
371
nature could be an isolated continental crustal block (microcontinent) or volcanic features
372
associated with mantle plumes. In addition, both maps show a roughly N-S-oriented abyssal
373
valley between the two RGR segments (gray polygon in Figure 4B). Its curved shape to the east
374
is coincident with the dextral movement of the E-W trending fracture zones that cross the whole
375
RGR. In this valley, isolated residual negative anomalies with values between -3 and -16 mGal
376
(located in the blue outlines in Figure 4B) are associated with seamounts and guyots that
377
probably are composed by magmatic rocks with lower density than the host oceanic crust. Local
378
gravity lows suggest the presence of areas with significant sedimentary accumulations in the
379
RGR, which formed interlayered with volcanic structures (red outlines in Figure 4B). The
380
formation of the abyssal valley is associated with a major change in the rotation pole and the
AC C
EP
TE D
M AN U
366
ACCEPTED MANUSCRIPT 18
381
rearrangement of the plates at around 84 Ma, which will be discussed in the simplified
382
temporal/spatial reconstruction of the RGR in the next chapter. In the central portion and flanks of the RGR, residual gravity lows (-7 and -3 mGal; white
RI PT
383
polygons in Fig. 5A) reveal the main depocenters of the Maastrichtian to Pleistocene volcano-
385
sedimentary sequences, which overlay the Santonian-Coniacian volcanic basement (ca. 82-96
386
Ma) drilled in the borehole 516F (Gamboa and Rabinowitz, 1984; Mohriak et al., 2010). In
387
addition, two 1,200-km-long NW-SE-oriented negative anomalies cross the WRGR and ERGR.
388
They are associated with the Cruzeiro do Sul Rift (CSR) and the thick sedimentary infill
389
deposited within the trough (white dashed lines in Figure 5A). The uplifted footwall blocks form
390
considerable local relieves across the rift-bounding escarpment and are capped by a narrower
391
sedimentary sequence than the adjacent areas, as was described by Mohriak et al. (2010). This
392
region corresponds to a set of positive residual Bouguer anomalies with a mean value of 10 mGal
393
(pink polygons in Figure 5A). This gravity pattern on the flanks also indicates an aborted rift that
394
was divided into smaller compartments along its length.
M AN U
TE D
The Cruzeiro do Sul Rift (CSR) is also well defined by two extensive negative free-air
EP
395
SC
384
gravity anomalies (pink dashed line in Figure 5B) of -31 and -141 mGal in the WRGR and
397
ERGR, respectively. Extensive free-air gravity lows occur along the CSR. In the ERGR, these
398
free-air gravity lows reach values up to -150 mGal in the deepest part of the rift (~ 5,800 m). In
399
contrast, the rift escarpments are marked by positive free-air anomalies between 59 and 85 mGal
400
(Figures 5B and 6).
401
AC C
396
Extensive E-W gravity lineaments are observed on the northern, southern and eastern
402
boundaries of the RGR (black dashed line in Figure 5B). This anomalous pattern is represented
403
by alternating positive and negative free-air anomalies associated with the FZs and the structural
ACCEPTED MANUSCRIPT 19
highs and lows of the basement. The rise of the mantle material and its rapid cooling caused
405
topographic elevation on the walls of the FZs. These structural highs may also be associated with
406
vertical tectonism of crustal and upper mantle blocks (Bonnati, 1978), and the valleys probably
407
resulted from the stress regime due to transcurrent movements in the active segments of these
408
fractures (Peive, 2006) near the RGR.
Profiles A-A' and B-B' show the correlation between the Bouguer anomalies, free-air
SC
409
RI PT
404
anomalies, and the bathymetry of the RGR (Figure 6). Profile A-A' crosses the WRGR and the
411
northern portion of the ERGR, whereas Profile B-B' crosses only the southern portion of the
412
ERGR. The lowest amplitudes of the Bouguer anomalies suggest that the Moho is deeper beneath
413
the WRGR and ERGR, especially between distances of 500 and 900 km in Profile A-A' and 400
414
and 750 km in Profile B-B’. In both profiles, gravity minima are observed in the CSR region,
415
whose central valley rose in response to the mechanical subsidence that occurred along this
416
structure when extensional forces divided the RGR during the Cenozoic. Short wavelength free-
417
air positive and negative anomalies show a series of seamounts and guyots near the RGR in the
418
NE portion of Profile A-A' and in the SW and NE portions of Profile B-B'. These volcanic
419
features are typical of oceanic crust affected by hot spots or anomalous magmatism along the
420
fracture zones.
TE D
EP
AC C
421
M AN U
410
Finally, a semicircular positive free-air anomaly (on the order of 20 mGal) is located in
422
the southern portion of the WRGR. This may be associated with a possible volcanic massif that is
423
composed of a set of volcanic cones (white circle in Figure 7). These massifs are associated with
424
volcanic eruptions in the ocean and are morphologically distinct from seamounts; they are
425
generally much larger and form domes with gentle slopes, whereas seamounts are typically high
426
with steeper slopes (Sager et al., 2013; Zhang et al., 2015). The randomly distributed volcanic
ACCEPTED MANUSCRIPT 20
cones imply that secondary eruptions may have occurred on this massif. Zhang et al. (2016)
428
described a similar anomalous pattern on the massif in the Shatsky Rise on the eastern margin of
429
Japan.
430 431
5.2. Magnetic Signatures in the RGR region
433
SC
432
RI PT
427
The typical reversal pattern of marine magnetic anomalies is highlighted on the RTP, TDR and HGA maps (Figures 8 and 9), which record the polarity reversals of the Earth's
435
magnetic field while the oceanic crust was being produced at mid-ocean ridges. Linear and
436
alternating positive and negative anomalies are oriented in the north-south direction and crosscut
437
by E-W trending magnetic lineaments related to fracture zones. The long magnetic stripes are
438
sub-parallel to the South American continental margin and roughly follow their inflections. The
439
magnetic anomalies present a significant inflection to the northeast in the region west of the
440
WRGR to the eastern limit of the ERGR, between the fracture zones Meteor and Rio Grande
441
(Figure 8). The magnetic anomaly related to Isochrone C34 (84 Ma) stands out. It is represented
442
by an extensive, well-defined negative magnetic anomaly that crosses the two segments of the
443
RGR (Figure 8). West of Isochrone C34, the magnetic stripes are wider with lower amplitudes,
444
except for the region encompassing the São Paulo Plateau and the WRGR, which present high
445
amplitudes of the analytical signal (Figure 8B). East of Isochrone C34, these lineaments are
446
oriented N-S, secondary NE-SW and NW-SE. This isochrone characterizes the end of the
447
Cretaceous magnetic quiescence period, during which the frequency of the geomagnetic reversals
448
declined steadily to its lowest point (without reversals) at approximately 84 Ma (Cox, 1973).
449
After a magnetically quiet period between 120 and 84 Ma (Malinverno et al., 2012), the
AC C
EP
TE D
M AN U
434
ACCEPTED MANUSCRIPT 21
450
frequency of these reversals increased continuously (Cox, 1973), and the magnetic anomalies
451
were better defined. In addition, a series of E-W-oriented magnetic lineaments that cross the NNE-SSW
RI PT
452
magnetic anomalies was observed (Figures 9 and 10). These lineaments are associated with FZs
454
and their continuations in the rise region, which are not easily observed in the gravity and
455
bathymetric data. The FZs compartmentalize the RGR and are scars produced by transform faults
456
that cross the mid-ocean ridge, deforming the oceanic crust throughout its extent (Adams, 2002).
457
The mapping of these structures, which are hundreds of kilometers long, is essential for
458
understanding the structural framework of the South Atlantic and, in particular, of the RGR,
459
which is transversally crossed by many of these fracture zones. Approximately eleven major FZs
460
are located between 28.7°S and 35.3°S latitude, and some were not identified in earlier studies,
461
such as those at 28.7°S, 29.2°S, 31.1°S, 31.6°S, 32.1°S, 33.7°S and 35.3°S. Additionally, other
462
FZs were better mapped in the present study, such as the Chuí and Cox FZs (Figures 9D and 10).
463
The nomenclature of the new discovered FZs was defined based on the current latitudes of the
464
expressive lineaments of these FZs.
465
EP
TE D
M AN U
SC
453
6. DISCUSSIONS
467
6.1 GEOTECTONIC EVOLUTION OF THE RIO GRAND RISE (100 - 55 Ma)
468 469
AC C
466
A kinematic reconstruction of the temporal and spatial evolution of the RGR and the Mid-
470
Atlantic Ridge segments was performed based on bathymetric, magnetic and free-air and
471
Bouguer gravity data (Figures 11 to 15). In addition, Figure 10 shows a simplified version of the
472
evolution of the RGR in five intervals during this period based on reversal magnetic pattern
ACCEPTED MANUSCRIPT 22
(isochrons described by Muller et al., 2008) and E-W trending magnetic lineaments (fracture
474
zones). Our plate reconstruction is quite similar with the most recent models (e.g., Moulin et al.,
475
2010; Heine et al., 2013; Pérez-Díaz and Eagles, 2014), with the advantages of being the first to
476
integrate bathymetric, gravity (free-air and Bouguer) and magnetic data and to focus on the
477
region of the RGR.
The plate reconstruction maps of South Atlantic evolution at intervals of 5 My support the
SC
478
RI PT
473
hypothesis that the RGR formed due to the excess magmatism from the Tristan da Cunha Plume,
480
which was located between the Rio Grande (North) and 35.3°S (South) fracture zones (FZs) for
481
approximately 45 Ma between late Cretaceous and Paleocene (ca. 95 - 60 Ma). The existence of a
482
microcontinent at the base of the RGR, as suggested by Stica et al. (2014), does not seem to be
483
supported by the present model. This hypothesis imply that the RGR would be located in the
484
African plate at the onset of the seafloor spreading (Barremian; ~127 Ma) and at some point in
485
the South Atlantic growth, the spreading axis jumped eastwards, leaving behind an abandoned
486
mid-oceanic ridge. Pérez-Díaz and Eagles (2014) interpreted the Vema Channel as an abandoned
487
mid-ocean ridge segment in the South American Plate south of the RGR based only on free-air
488
gravity anomalies. However, there is no evidence of the reversal pattern of magnetic anomalies
489
(Figures 8 to 10) favoring this model. Furthermore, these authors consider the RGR a volcanic
490
bulge owing to the action of the Tristan da Cunha plume, which seems inconsistent with a
491
spreading axis supposed surrounded by continental segments.
TE D
EP
AC C
492
M AN U
479
Nunns (1983) depicted the typical distribution of reversal magnetic pattern of marine
493
anomalies in the North Atlantic Ocean. According to this author, in the Norwegian-Greenland
494
Sea, the Jan Mayen microcontinent, with chaotic magnetic pattern, is surrounded by the
495
abandoned Aegir Ridge to the east (magnetic anomalies 24 to 12) and the active Kolbeinsey
ACCEPTED MANUSCRIPT 23
Ridge to the west (magnetic anomalies 7 to 0). Instead, the magnetic anomalies in the South
497
Atlantic plate, comprising the RGR and the Vema Channel, have a roughly north-south trending
498
distribution bended eastwards at the RGR and symmetric to its counterpart in the African plate.
499
In addition, the absence of symmetric magnetic stripes could indicate another scenario in which
500
the RGR represent a continental ribbon and the Vema Channel an aborted rift, regarding to the
501
model of rifting evolution developed by Péron-Pinvidic and Manatschal (2010) for the North
502
Atlantic. In this case, no oceanic crust would have been created between the São Paulo Ridge and
503
the RGR. However, this model contradicts the oceanic nature of the crust, whose boundary with
504
the continental crust was mapped 250 km to the west in ultra-deep seismic sections by Zalán et
505
al. (2011).
M AN U
SC
RI PT
496
Ussami et al. (2012) discussed the possibility of the RGR being a microcontinent
507
regarding to its 30-km crustal thickness. According to them, essential tectono-magmatic
508
conditions, such as re-rifting of a young continental margin and asymmetric sea-floor spreading,
509
are lacking. On the other hand, flexural analysis of free-air gravity data indicates a construction
510
of the RGR related to magmatic episodes. And thus continental crust slivers, such as those
511
recently discovery (Fioravanti, 2014), may have been raised to the surface during the magma
512
emplacement.
EP
AC C
513
TE D
506
During the interval between the 100 and 90 Ma ages (Figures 11 and 12), the center of the
514
hotspot was initially located in the region of the mid-ocean ridge segments between the Rio
515
Grande and Porto Alegre FZs during the Cenomanian to Turonian. The segments north and south
516
of the hotspot migrated northwest, and the segments to the southwest, between the Porto Alegre
517
and Chui FZs, appear to have remained fixed but rotated slightly NNE from their original NE-
518
SW orientation. Starting at 95 Ma, the western portion of the RGR began to form between the
ACCEPTED MANUSCRIPT 24
28.7°S and Porto Alegre FZs and gradually developed southwards to the 31.6°S FZ at
520
approximately 90 Ma. The effect of the formation of the aseismic ridge is illustrated by a
521
semicircular Bouguer gravity anomaly low of 270 mGal in regions of bathymetric high that
522
reaches 400 m at the ends of this rise segment (Fig. 11B).
523
RI PT
519
Between the 90 and 80 Ma ages (Figures 12 and 13), the mid-ocean ridge segments
affected by the hotspot magmatism are distributed between the Rio Grande and 32.1°S FZs. On
525
the South American plate, the entire WRGR developed until around 80 Ma (Campanian). Prior to
526
that, the northern portion of the ERGR began to form between the Rio Grande and 29.2°S FZs
527
starting at ca. 85 Ma (Coniacian-Santonian), which is suggested by the presence of an E-W-
528
oriented gravity anomaly minimum of ca. 370 mGal (A in Figure 12B). At ca. 80 Ma
529
(Campanian), a semicircular valley (gray polygon in Figure 13A) apparently formed from north
530
to south on the ERGR-Walvis plateau on the mid-ocean ridge axis between the 29.2°S and Porto
531
Alegre FZs (yellow dashed line in Figure 13A). As noted by Gente et al. (2003) and Pessanha
532
(2011), this valley represents a plateau rupture related to a period of varying magmatic
533
contributions from the Tristan da Cunha hotspot. The local magmatism decrease differs from
534
what occurs in the Azores, where the increasing distance between the hotspot and ridge axes is
535
the main cause of the decrease in magma supply and of the oceanic plateau break-up. The
536
reconstruction plate model reinforces that during the evolution of the RGR-Walvis system in this
537
period, the main cause of the RGR-Walvis plateau break-up in this area was probably the
538
decrease in the magma supply in these ridge segments. In addition, the decreased magmatic
539
contribution from the plume in the ridge segments between the 29.2°S and 31.1°S FZs is also
540
indicated by the high gravity anomaly (on the order of 460 mGal), which represents a possible
541
area of thinner crust (gray polygon on the Bouguer map in Figure 13B).
AC C
EP
TE D
M AN U
SC
524
ACCEPTED MANUSCRIPT 25
542
After a major change of the rotational pole of the South American and African plates at 84 Ma (Muller et al., 1997) and its consequent spatial rearrangement, the period between 80 and 70
544
Ma (Figures 13 and 14) is represented by a striking instability of the accretion system of the mid-
545
ocean ridge axes in Campanian to Maastrichtian (Pessanha, 2011). Due to the displacements of
546
the South American and African plates during this period, the plume, which was initially
547
positioned over the ridge axis and impacts the region between the Rio Grande and 31.1°S FZs,
548
began to move under the African plate. However, it continued to provide magmatic material to
549
the ERGR between the 28.7°S and Porto Alegre FZs until approximately 75 Ma and extended the
550
magmatism to the 33.7°S FZ at around 70 Ma. The repositioning of the mid-ocean ridge
551
segments between the Porto Alegre and Chui FZs closer to the hotspot becomes evident during
552
this period in the evolutionary reconstruction of the RGR, as shown between intervals II and III
553
of Figure 10. These segments maintained a NNE-SSW orientation but rotated significantly
554
counterclockwise in relation to geographic north, while the segments were rearranged, and new
555
seafloor formed. Finally, the topographic plain formed between the WRGR and ERGR between
556
80 and 70 Ma (Campanian to Maastrichtian) due to the combination of the rearrangement of the
557
lithospheric plates and the migration of ridge segments closer to the hotspot during this period.
558
This reconfiguration occurred in two regions, one between the Rio Grande and Porto Alegre FZs
559
and another to the south between the Porto Alegre and Montevideo FZs. According to Muller et
560
al. (2008), these regions have different spreading rates (≈ 30 and 110 km/My, respectively) and
561
asymmetries (≈ 60% and 90%, respectively). The difference between these rates explains the E-
562
W extent in the southern portion of the plain.
563 564
AC C
EP
TE D
M AN U
SC
RI PT
543
Between 70 and 60 Ma (Figures 14 and 15), the ridge segments between the 31.1°S and Cox FZs were rearranged near the hotspot with a predominant N-S orientation in Maastrichtian to
ACCEPTED MANUSCRIPT 26
Paleocene. During the 65 Ma period, the north and central portions of the ERGR segment had
566
already formed, and the influence of the plume on the ridge segments between Cox and 35.3°S
567
FZs resulted in the development of the southern portion of the ERGR. During the same period,
568
the Chuí and 32.1°S FZs may have stopped developing, or their signatures were not clear in the
569
magnetic data (Figures 8 and 9).
Finally, the southern portion of the ERGR developed until approximately 55 Ma (Figure
SC
570
RI PT
565
15B). During this period (Eocene), the plume was located under the African plate, which reduced
572
its influence on the South American plate side, and it was limited to forming isolated seamounts.
573
However, intense volcanic activity continued in the African plate, which generated the western
574
portions of the Walvis Ridge. The presence of a positive Bouguer gravity anomalies on the order
575
of 470 mGal (gray polygon in Figure 15D) suggests that the oceanic crust in this region of the
576
South American plate has the normal thickness (~ 10 km) reflecting the absence of the magmatic
577
influence of the Tristan da Cunha Plume.
TE D
M AN U
571
578
6.2 TECTONIC SUBDIVISION OF THE CRUZEIRO DO SUL RIFT
EP
579 580
The aborted Cruzeiro do Sul Rift (CSR) is well defined in the gravity data by extensive
AC C
581 582
negative Bouguer and free-air anomalies that follow the NW-SE axis of the central valley
583
(Figures 5 and 6). The flanks of the rift valley progressively decrease towards the southeastern
584
portions of the rift followed by a decrease in the gradient of the free-air anomalies (Figure 7),
585
which indicates that subsidence along this structure progressively increased in this direction. In
586
contrast, the magnetic signature of the rift is not well defined in the dataset used in this study
587
(Figure 8).
ACCEPTED MANUSCRIPT 27
The CSR developed in an extensive deformation zone that cuts the RGR at a high angle as
589
suggested by Mohriak et al. (2010). The aborted rift region represents an area that accommodated
590
the deformation of the extensional faults of the South Atlantic rifting. These movements between
591
large crustal blocks in the accommodation areas of the southeastern margin of the South
592
American Platform played a fundamental role on the magmatic history and development of the
593
Santos and Campos marginal basins, both prior to and after breakup (Mohriak et al., 2010; Quirk
594
et al., 2013).
SC
The breakup and seafloor spreading of the South Atlantic Ocean are a combination of
M AN U
595
RI PT
588
extensional tectonics and dextral shear movement associated with transform faults between the
597
South American and African plate segments that include the RGR (Nurnberg and Muller, 1991;
598
Mohriak et al., 2010). Therefore, the transfer zone in which the CSR developed had an NE-SW
599
oriented extensional component that is oblique to the regional dextral strike-slip movements of E-
600
W trending fracture zones, which affected its development (Figure 16A). These forces along the
601
FZs were possibly activate at the onset of the rift in Barremian-Aptian (Mohriak, 2010) and after
602
during the seafloor fragmented the CSR in large segments between the Paleogene and Neogene
603
ages (Figure 17). According to Cobbold (2001), the entire southeastern margin of South America
604
was reactivated by dextral and sinistral strike-slip movements along the extensional faults of the
605
South Atlantic opening and some segments of transfer faults near the Brazilian coast. These
606
reactivations occurred during the Upper Cretaceous, Eocene and Neogene (Cobbold, 2001) and
607
are correlated with changes in the convergence conditions (direction and speed) of the Andean
608
margin of the South American continent. The tectonic reactivations during the Eocene and
609
Neogene affected the region containing the Cenozoic rift basins along the southeastern margin in
610
Brazil (Zalán and Oliveira, 2005). As well as deeper areas of the South Atlantic are also
AC C
EP
TE D
596
ACCEPTED MANUSCRIPT 28
deformed by these tectonic events, as indicated by the prominent inflections observed along the
612
CSR that developed over the RGR (I and II in Figure 16A). These inflections coincide with areas
613
of FZs and might represent reactivations of old pre-existing transform faults.
614
RI PT
611
The CSR was divided into two segments mainly by the combination of the transcurrent movements between the FZs and the transfer zone in which the rift formed. The northern segment
616
(NCSR) crosses the northwestern portion of the RGR, and the southern segment (SCSR) is
617
located in the southern region of the eastern portion of the rise (Figure 16B). Both segments were
618
rotated counterclockwise in response to the dextral transcurrent forces of the 29.2°S, Cox and
619
Meteor FZs. In the South Atlantic oceanic crust, the NW-SE orientation of the continental shear
620
zones of the Precambrian basement grain is realigned with the E-W orientation of the FZs
621
(Mohriak, 2003), and this alignment is observed in both segments of the rift. Part of the NCSR
622
that formed in the accommodation zone becomes parallel to the Cox FZ at 33.1°S latitude and
623
passes through the eastern portion of the RGR (Fig. 16). In turn, the SCSR segment becomes
624
parallel to the Meteor FZ at 34.5°S latitude.
M AN U
TE D
In addition, the NCSR and SCSR can be subdivided into smaller compartments, which
EP
625
SC
615
were also rotated counterclockwise by strike-slip movements of the FZs. The WRGR and ERGR
627
consist of four and three compartments, respectively (I and II in Figure 17). The compartment
628
between the Porto Alegre and 31.1°S FZs had the greatest rotation along the CRS. In this
629
compartment, a portion of the rift was shifted by a sinistral strike-slip force along a NE-SW-
630
oriented structure (III in Figure 17). This structure is well defined in the TDR map (A in Figure
631
17 III), and it may be related to an extensional fault that formed or was reactivated during the
632
opening of the CRS or an old ridge axis that was reactivated during the development of the rift.
633
AC C
626
ACCEPTED MANUSCRIPT 29
634
6.3 CURVATURE OF THE SOUTHERN PORTION OF THE JEAN CHARCOT CHAIN
635
Similar to the Cruzeiro do Sul Rift (CSR), the Jean Charcot seamount chain (JCC) is not
637
well defined in the magnetic data. However, these structures appear generally well-defined as a
638
series of short-wavelength positive free-air gravity anomalies oriented in NW-SE and NE-SW
639
directions in the northern and southern segments, respectively (Figure 18). They are associated
640
with around 1,000 km long semicircular volcanic structures identified in the bathymetric data
641
(Figure 2). It extends from the Brazilian continental margin for approximately 500 km to the
642
RGR, where it curves sharply SW for another 500 km. Many questions remain about the
643
seamounts on the curved segment, especially about their origin. They are oriented NE-SW, which
644
is opposite to the NW-SE trend of the northern segment of the chain (Figure 2).
M AN U
SC
RI PT
636
One explanation for the differences between the trends of the Jean Charcot Chain
646
segments can be extracted from an analogy with several magmatic events that occurred along the
647
southeastern margin of Brazil. Analyses of seismic sections by Oreiro et al. (2006) indicate that
648
the post-salt magmatic events on the southeastern margin of the South American Platform were
649
the result of the reactivation of deep fault zones, many of which formed during the break-up of
650
the Gondwana supercontinent and by changes in the rotation pole of the South American plate
651
between 84 to 50 Ma. These magmatic events occurred near the intersections between NW-SE-
652
oriented fault zones (strike-slip faults) and NE-SW-oriented fault zones (normal faults) and along
653
the lineament formed by the Poços de Caldas – Cabo Frio alkaline massifs in southeastern Brazil.
654
These data suggest that the magmatism did not occur due to the movement of the plate over a
655
mantle plume because the reactivated faults (mainly the NW-SE-oriented strike-slip faults)
AC C
EP
TE D
645
ACCEPTED MANUSCRIPT 30
656
appear to have cut the entire lithosphere, reached the asthenosphere and caused partial melting of
657
the mantle by pressure relief. A similar situation may have occurred in the southern portion of the Jean Charcot Chain
RI PT
658
but with a peculiar feature. The seamounts in this chain are located with or very close to the FZs,
660
which cross this area, without however presenting a clear alignment whit these E-W trending
661
structures (Figure 18A). On the other hand, the series of seamounts roughly coincides with the
662
magnetic lineaments, especially between Meteor and Rio Grande FZs (Figures 18B and C). The
663
intersections between these reactivated FZs and a likely NE-SW-oriented region of extensional
664
deformation, which developed or was also reactivated during the opening of the CSR, may have
665
formed a large zone of crustal weakness (gray dashed line in Figures 18A and B). In this context,
666
these intersections must have served as conduits for the magmatic rises that formed the southern
667
portion of the mountain range. However, the Jean Charcot Chain and the distribution of its
668
seamounts still require additional detailed studies.
669
671
9. CONCLUSIONS
EP
670
TE D
M AN U
SC
659
The application of enhancement techniques to gravity and magnetic anomalies from
673
global potential models allowed new regional geophysical mapping of the Rio Grande Rise
674
(RGR) in the South Atlantic. The regional Bouguer low indicates a thicker crust related to normal
675
oceanic crust in the adjacent Brazil and Argentina oceanic basins. A crustal root (microcontinent
676
or oceanic origin) beneath the RGR is expected in response to the effect of isostatic compensation
677
of the 2000-m height large elliptical bulge. Furthermore, residual Bouguer lows allow dividing
678
the RGR in two segments (WRGR and ERGR) and indicate regions associated with thick
AC C
672
ACCEPTED MANUSCRIPT 31
sedimentary packages that deposited over the RGR. Finally, the Cruzeiro do Sul Rift (CSR) was
680
well defined by two extensive NW-SE-oriented negative Bouguer and free-air anomalies across
681
the WRGR and ERGR.
682
RI PT
679
The new magnetic maps show the typical reversal magnetic pattern of the oceanic crust, except in the RGR, where the magnetic response is more complex and the N-S orientation of the
684
magnetic anomalies markedly bend to NE-SW. Magnetic anomalies also allowed precisely
685
mapping of FZs throughout the rise region. Approximately eleven FZs observed in the magnetic
686
data compartmentalized the entire rise and the CSR. Based on the processed magnetic anomalies
687
and magnetic isochrones, we identified the spreading pattern of the oceanic crust and its temporal
688
relationship with the formation of the RGR during the Albian to Eocene (~100-55 Ma). In face to
689
the finding of continental rocks recently dredged on the rise, speculation about the RGR being a
690
microcontinent has been renewed. However, the reversal magnetic pattern unravels no evidence
691
of aborted spreading center, indicative of ridge jumps and microcontinent separation. An
692
alternative explanation is that slivers of continental blocks rose during the magma emplacement.
M AN U
TE D
The temporal-spatial evolution of the RGR, Walvis Ridge and Tristan da Cunha Plume
EP
693
SC
683
was analyzed with a plate reconstruction model. The proposed model integrates gravity, magnetic
695
and bathymetric data and covers the interval from 100 to 55 Ma (Albian to Eocene ages), which
696
represents a period of rapid expansion of the seafloor in the South Atlantic. Our reconstruction
697
model is consistent with the modern models of the South Atlantic opening, but it is the only one
698
to specifically focus on the RGR evolution. The proposed model supports the idea that the
699
rearrangement of the segments of the Mid-Atlantic Ridge near the Tristan da Cunha hotspot
700
combined with the additional supply of magma from the hotspot generated the anomalously high
701
and semicircular morphology of the RGR. The extensive FZs may have formed barriers or
AC C
694
ACCEPTED MANUSCRIPT 32
702
conduits for the magmatism from the interaction of the plume with the segments of the mid-ocean
703
ridge during the formation of the RGR. In addition, the FZs, revealed by magnetic lienaments, coincide with the observed
RI PT
704
inflections along the CSR. These inflections might represent pre-existing transform fault zones
706
that were reactivated during the tectonic-magmatic event that formed the aborted rift (Paleogene-
707
Neogene). The division of the rift into two segments (NCSR and SCSR) was based on the
708
significant offsets along the rift. Other smaller counterclockwise-rotated compartments were also
709
observed in the NCSR and SCSR and represent a response to strike-slip dextral movements in the
710
FZs combined with a NE-SW trending extensional regime, which was responsible for the rift
711
formation in the RGR.
M AN U
SC
705
The gravity and magnetic maps also shed light on the origin of the Jean Charcot Chain.
712
The alignment of the series of seamounts, mainly in the southern segment and part of the northern
714
segment to the Rio Grande FZ, with magnetic anomalies suggests that the mechanism of
715
emplacement of these magmatic bodies was influenced by the seafloor spreading of the mid-
716
Atlantic Ridge. This mechanism was still strongly controlled by FZs, which strongly affect this
717
region.
719 720
EP
AC C
718
TE D
713
ACKNOWLEDGMENTS
This research was supported by the IODP/CAPES Program [Project n. 13/2014]. The
721
authors thank the International Gravimetric Bureau (IBG), Commission for the Geological Maps
722
of the World (CCGM) and National Oceanic and Atmospheric Administration (NOAA) for
723
geophysical and bathymetric datasets used in this study. The authors are grateful to the two
ACCEPTED MANUSCRIPT 33
724
anonymous reviewers, whose comments significantly improved the manuscript. ILGG and DLC
725
thank the PRH-PB 229 Scholarship Program (Petrobras) and CNPq for their respective grants.
RI PT
726
REFERENCES
728
Abdelsalam, M. G., Katumwehe, A. B., Atekwana, E. A., Le Pera, A. K., Achang, M. 2016. The
729
Paleoproterozoic Singo granite in the South-Central Uganda revelead as a nested igneous ring
730
complex using geophysical data. Journal of African Earth Sciences, 116, p. 198-212.
SC
727
M AN U
731 732
Adam, C., Vidal, V., Escartin, J. 2007. 80-myr history of buoyancy and volcanic fluxes along the
733
trails of the Walvis and St. Helena hotspots (South Atlantic). Earth and Planetary Science Letters,
734
261(3-4), p. 432-442.
735
Alves, E.C. 1981. Estruturas da margem continental sul brasileira e das áreas oceânicas e
737
continentais, adjacentes. In: PETROBRAS. Estruturas e tectonismo da margem continental
738
brasileira, e suas implicações nos processos sedimentares e na avaliação do potencial de recursos
739
minerais; relatório final. Rio de Janeiro, CENPES/DINTEP. Projeto REMAC 9, p. 145-170.
EP
AC C
740
TE D
736
741
Alves, E. C. 2002. Zonas de fraturas oceânicas e suas relações com a compartimentação tectônica
742
do sudeste do Brasil. Tese de Doutorado. Departamento de Geologia da Universidade Federa do
743
Rio de Janeiro, 247 p.
744 745
Amante, C. and B. W. Eakins. 2009. ETOPO1 1 Arc-Minute Global Relief Model: Procedures,
746
Data Sources and Analysis. NOAA Technical Memorandum NESDIS NGDC-24, March, 19 p.
ACCEPTED MANUSCRIPT 34
747
Asmus, H. E. 1975. Controle estrutural da deposição mesozóica nas bacias da margem
749
continental brasileira. Rev. Bras. Geoci., 5, p. 160–174.
RI PT
748
750
Asmus, H. E., Ferrari, A. L. 1978. Hipótese sobre a causa do tectonismo cenozóico da região
752
Sudeste do Brasil. Rio de Janeiro: PETROBRAS. Série Projeto REMAC 4, p.75–88.
SC
751
753
Asmus, H. E. 1984. Geologia da margem continental brasileira. In: SCHOBBENHAUS, C.;
755
CAMPOS, D.A.; DERZE, G.R.; ASMUS, H.E. coords. Geologia do Brasil. Texto explicativo do
756
mapa geológico do Brasil e da área oceânica adjacente incluindo depósitos minerais. Brasília:
757
DNPM, p. 443–472.
M AN U
754
758
Balmino, G., Vales, N., Bonvalot, S. and Briais, A., 2012. Spherical harmonic modeling to ultra-
760
high degree of Bouguer and isostatic anomalies. Journal of Geodesy. July 2012, Volume 86,
761
Issue 7, p. 499-520.
EP
762
TE D
759
Baker, P. F., Carlson, R. L., Johnson, D. A., Cepek, P., Coulbourn, W., Gamboa, L. A.,
764
Hamilton, N., Melo, U., Pujol, C., Shor, A. N., Suzyumov, A. E., Tjalsma, L. R. C., Walton, W.
765
H. and Weiss, W. 1981. Deep Sea Drilling Project Leg 72: Southwest Atlantic paleocirculation
766
and Rio Grande Rise tectonics. Geol. Soc. Am. Bull., 92: p. 294–309.
767
AC C
763
ACCEPTED MANUSCRIPT 35
Baker, P. F. 1983. Tectonic evolution and subsidence history of the Rio Grande Rise. In: Baker,
769
P. F., Carlson, R. L. et al. (eds) Initial Reports of the Deep Sea Drilling Project, Volume 72. US
770
Government Printing Office, Washington, DC, p. 953–976.
RI PT
768
771
Blakely, R.J., 1996. Potential theory in gravity and magnetic applications. 2nd ed. Cambridge
773
University Press, Cambridge (UK), 441 p.
SC
772
774
Bonatti, E. 1978. Vertical Tectonism in Oceanic Fracture Zones. Earth Planet. Sci. Lett., 37, p.
776
369-379
M AN U
775
777
Bonvalot, S., Balmino, G., Briais, A., Kuhn, M., Peyrefitte, A., Vales, N., Biancale, R., Gabalda,
779
G., Reinquin, F., Sarrailh, M., 2012. World Gravity Map. Commission for the Geological Map of
780
the World. Eds. BGI-CGMW-CNES-IRD, Paris, 8 p.
TE D
778
781
Boyden, J. A., Müller, R. D., Gurnis, M., Torsvik, T. H., Clark, J. A., Turner, M., Ivey-Law, H.,
783
Watson, R. J., and Cannon, J. S. 2011. Next-generation plate-tectonic reconstructions using
784
GPlates. in: Geoinformatics: Cyberinfrastructure for the Solid Earth, edited by: Keller, G. R. and
785
Baru, C., Cambridge University Press, Cambridge, p. 95–114.
AC C
786
EP
782
787
Bryan, P.C., Cherkis, N.Z. 1995. The Bahia Seamounts: test of a hotspot model and a preliminary
788
South American Late Cretaceous to Tertiary Apparent Polar Wander Path. Tectonophysics 241,
789
p. 317–340.
790
ACCEPTED MANUSCRIPT 36
791
Cainelli, C. & Mohriak, W.U. 1999. Some remarks on the evolution of sedimentary basins along
792
the Eastern Brazilian continental margin. Episodes, vol. 22, n. 3, p. 206–216.
RI PT
793
Chang, J-H., Hsieh, H-H., Mirza, A., Chang, S-P., Hsu, H-H., Liu, C-S., Su, C-C., Chiu, S-D.,
795
Ma, Y-F., Chiu, Y-H., Hung, H-T., Lin, Y-C., Chiu, C-H. 2016. Crustal structure north of the Itu
796
Aba Island (Taiping Island), southern margin of the South China Sea. Journal of Asian Earth
797
Sciences, No prelo.
SC
794
M AN U
798 799
Chian, D., Reid, I.D. & Jackson, H.R. 2001. Crustal structure beneath Orphan Basin and
800
implications for nonvolcanic continental rifting. Journal of Geophysical Research, 106 (B6), p.
801
10923–10940.
802
Cobbold, P. R.; Meisling, K. E.; Mount, V. S. 2001. Reactivation of an obliquely rift-ed margin,
804
Campos and Santos basins, southeastern Brazil. Am. Ass. Petr. Geol. Bull., Tulsa, Okla., v. 85, n.
805
11, p. 1925–1944.
EP
806
TE D
803
Cox, A. 1973. Plate tectonics and geomagnetic reversals. San Francisco, California: W. H.
808
Freeman. pp. 138–145, 222–228.
809
AC C
807
810
De Castro, D.L., Medeiros W.E., Jardim de Sá E.F. & Moreira J.A.M. 1998. Mapa gravimétrico
811
do Nordeste Setentrional do Brasil e margem continental adjacente: interpretação com base na
812
hipótese de isostasia. Rev. Bras. de Geof., 16 (2/3), 115–131.
813
ACCEPTED MANUSCRIPT 37
De Castro, D. L., Bezerra, F. H. R., Sousa, M. O. L., and Fuck, R. A., 2012. Influence of
815
Neoproterozoic tectonic fabric on the origin of the Potiguar Basin, northeastern Brazil and its
816
links with West Africa based on Gravity and Magnetic Data, Journal of Geodynamics, 54, 29–42.
RI PT
814
817
Detrick, R. S., Sclater, J. G. and Thiede, J. 1977. The subsidence of aseismic ridges. Earth and
819
Planetary Science Letters, 34, 185–196.
SC
818
820
Dietz, R.S. and Holden, J.C., 1970. Reconstruction of Pangea: breakup and dispersion of the
822
continents, Permian to present. J. Geophys. Res., 75, p.4939–4956.
M AN U
821
823
Estrella, G. O. 1972. O estágio rift nas bacias marginais do Leste Brasileiro. XXVI Congr. Bras.
825
Geol. da S.B.G., Belém-Pará, v.3, p. 29–34.
826
TE D
824
Fairhead, J. D., Marjorie, W. 2005. Plate tectonic processes in the south atlantic ocean: Do we
828
need deep mantle plumes? Special Paper - Geological Society of America, 388, p. 537–553.
829
EP
827
Fioravanti, C. 2014. Ecos da separação - Grandes blocos de rochas com idades e origens
831
diferentes se combinaram ao formar os dois lados do Atlântico Sul. Pesquisa Fapesp, 224, 58-61.
832
AC C
830
833
Fodor, R. V., Husler, J. W. & Kumar, N. 1977a. Petrology of volcanic rocks from an aseismic
834
rise: implications for the origin of the Rio Grande Rise, South Atlantic Ocean. Earth and
835
Planetary Science Letters, 35, p. 225–233.
836
ACCEPTED MANUSCRIPT 38
Fodor, R.V. and Thiede, J. 1977b. Volcanic breccia from DSDP Site 357: Implications for the
838
composition and origin of the Rio Grande Rise. In: P.R. Supko, K. Perch Nielsen et al., Initial
839
Reports of the Deep Sea Drilling Project, v. 39, U.S. Govt. Printing Office, Washington, D.C., p.
840
537–543.
RI PT
837
841
Fodor, R. V., Mckee, E. H., Asmus, H. E. 1984. K-Ar ages and the opening of the South Atlantic
843
Ocean: basaltic rocks from the Brazilian Margins. Marine Geology, 54 (1/2): M1–M8.
SC
842
M AN U
844 845
Fulfaro, V. J., Saad, A. R., Santos, M. V. and Vianna, R. B. 1982. Compartimentação e evolução
846
tectônica da Bacia do Paraná. Rev. Bras. Geoci, v. 12, n. 4, p. 590–611.
847
Gamboa, L. A. P. and Rabinowitz, P. D. 1981. The Rio Grande Fracture Zone in the western
849
South Atlantic and its tectonic implications. Earth Planet. Sci. Lett., 52, p. 410-418
TE D
848
850
Gamboa, L. A. P. and Rabinowitz, P. D. 1984. The evolution of the Rio Grande Rise inthe
852
southwest Atlantic Ocean. Mar. Geol. 58, p. 35–38.
EP
851
AC C
853 854
Gibson, S. A., Thompson, R. N, Day, J. A., Humphris, S. E and Dickins, A. P. 2005. Melt-
855
generation processes associated with the Tristan mantle plume: Constraints on the origin of EM-
856
1. Ear. Planet. Sci. Lett., v. 237, p. 744–767.
857 858
Heine, C., Zoethout, J., Müller, R. D. 2013. Kinematics of the South Atlantic rift. Solid Earth, 4,
859
215–253.
ACCEPTED MANUSCRIPT 39
860
Kaji, A. O., Guerra, J. V., Fernandes, A. M., De Oliveira, R. F., Silva, C. G. and Dos Reis, A. T.
862
2011. Potencial de transporte sedimentar pelas correntes de fundo na região do Canal de Vema
863
(Atlântico Sul). Revista Brasileira de Geofísica, 29(2), p. 385-400.
864
RI PT
861
Kodaira, S., Mjelde, R., Gunnarsson, K., Shiobara, H. & Shimamura, H. 1998. Structure of the
866
Jan Mayen microcontinent and implications for its evolution. Geophysical Journal International,
867
132, p. 383–400.
M AN U
SC
865
868 869
Leyden, R., Ludwig, W., Ewing, M. 1971. Structure of continental margin off Punta del Este,
870
Uruguay, and Rio de Janeiro, Brazil. Assoc. Petr. Geol. Bull., 55:2, p. 161–2.173.
871
Leyden, R. 1976. Salt distribution and crustal models for the eastern Brazilian margin. Anais
873
Acad. Bras. Ciênc., v. 48, p. 159–168.
EP
874
TE D
872
Macedo, J. M. 1989. Evolução tectônica da Bacia de Santos e áreas continentais adjacentes. Bol.
876
Geoc. Petrobrás, Rio de Janeiro, v. 3, no 3, p. 159–173.
877
AC C
875
878
Malinverno, A., Hildebrandt, J., Tominaga, M., Channell, J.E.T., 2012. M-sequence geo-
879
magnetic polarity time scale (MHTC12) that steadies global spreading rates and incorporates
880
astrochronology constraints. J. Geophys. Res.117, B06104.
881
ACCEPTED MANUSCRIPT 40
Maus, S., U. Barckhausen, H. Berkenbosch, N. Bournas, J. Brozena, V. Childers, F. Dostaler, J.
883
D. Fairhead, C. Finn, R. R. B. von Frese, C. Gaina, S. Golynsky, R. Kucks, H. Lühr, P. Milligan,
884
S. Mogren, R. D. Müller, O. Olesen, M. Pilkington, R. Saltus, B. Schreckenberger, E. Thébault,
885
and F. Caratori Tontini, 2009, EMAG2: A 2–arc min resolution Earth Magnetic Anomaly Grid
886
compiled from satellite, airborne, and marine magnetic measurements. Geochemistry,
887
Geophysics, Geosystems, 10, no. 8, 12 p.
SC
RI PT
882
888
Mitzusaki, A. M. P. 1986. Rochas igneo-básicas do Neocomiano da Bacia de Campos:
890
Caracterização e comportamento como reservatório de hidrocarbonetos. Dissertação de Mestrado.
891
Instituto de Geofísica, UFRJ, Rio de Janeiro. 104 p.
M AN U
889
892
894
Morgan, W. J., 1971. Convection plumes in the lower mantle. Nature, 230, p. 42–43.
TE D
893
Mohriak, W. U.; Rosendahl, B. R. 2003. Transform Zones In The South Atlantic Rifted
896
Continental Margins. Geological Society, London, Special Publications, 210, p. 211–228.
897
EP
895
Mohriak , W. U., Nóbrega, M., Odegard, M. E., Gomes, B. S., Dickson, W. G. 2010. Geological
899
and geophysical interpretation of the Rio Grande Rise, south-eastern Brazilian margin;
900
extensional tectonics and rifting of continental and oceanic crusts. Geological Society Publishing
901
House for EAGE (European Assoaciation of Geoscientists & Engineers), London, United
902
Kingdom, Stretching the crust and extending exploration frontiers, p. 231–245.
903
AC C
898
ACCEPTED MANUSCRIPT 41
904
Muller, R. D., Royer, G. S. 1997. A digital age map of the ocean floor. SIO Reference Series,
905
93:30.
RI PT
906
Muller R. D., Sdrolias M., Carmen Gaina Walter R. Roest. 2008. Age, spreading rates, and
908
spreading asymmetry of the world’s ocean crust. Geochemistry Geophysics Geosystems, 9 (4):
909
1–19.
SC
907
910
Nunns, A.G. 1983. Plate tectonic evolution of the Greenland-Scotland Ridge and surrounding
912
regions. In: Bott, M.H.P., Saxov, S., Talwani, M. and Thiede, J. (eds). Structure and
913
Development of the Greenland-Scotland Ridge: New Methods and Concepts. Plenum Press, New
914
York, 11-30.
M AN U
911
915
Nurnberg, D., Muller, R. D. 1991. Tectonic evolution of the South Atlantic from late Jurassic to
917
present. Tectonophysics, 191, 27–53.
EP
918
TE D
916
O’Connor, J. M. & Duncan, R. A. 1990. Evolution of the Walvis-Ridge-Rio Grande Rise hotspot
920
system: implications for African and South American plate motions over plumes. Journal of
921
Geophysical Research, 95 (17), 474-502.
922
AC C
919
923
O’Connor J. M., Le Roex A. P. 1992. South Atlantic hot spot-plume systems: distribution of
924
volcanism in time and space. Earth and Planetary Science Letters, 113 (3), 343–364.
925
ACCEPTED MANUSCRIPT 42
926
O'Connor, J. and Jokat, W. 2015. Tracking the Tristan-Gough mantle plume using discrete chains
927
of intraplate volcanic centers buried in the Walvis Ridge. Geology, 43 (8), p. 715-718.
RI PT
928 929
Ojeda, H. A. O., 1982. Structural framework, stratigraphy, and evolution of Brazilian marginal
930
basins. Am. Ass. Petr. Geol. Bull. v. 66, n. 6, p.732–749.
SC
931
Oreiro, S. G. 2006. Magmatismo e sedimentação em uma área na Plataforma Continental de
933
Cabo Frio, Rio de Janeiro, Brasil, no intervalo Cretáceo Superior – Terciário. Bol. Geoc.
934
Petrobras, Rio de Janeiro, v. 14, n.1, p. 95–112.
M AN U
932
935
Peive, A.A. 1976. Vertical Tectonic Movements of The Crust in Transform Fracture Zones of
937
The Central Atlantic. Geotectonics, 40 (1), 25-36.
938
TE D
936
Pérez-Díaz, L., and Eagles, G. 2014. Constraining South Atlantic growth with seafloor spreading
940
data. Tectonics, 33, 1848–1873.
941
EP
939
Péron-Pinvidic, G., and Manatschal, G. 2010. From microcontinents to extensional allochthons:
943
witnesses of how continents rift and break apart? Petroleum Geoscience, 16, 189–197.
944
AC C
942
945
Pessanha, I. B. M., 2011. Evolution Temporelle et Spatiale du Système d’Interaction entre le
946
point chaud de Tristan da Cunha et la Dorsale de l’Atlantique Sud. G´eophysique [physics.geo-
947
ph]. Université de Bretagne occidentale - Brest, Francais, 262 p.
948
ACCEPTED MANUSCRIPT 43
Phillips, J. D. 2001. Designing matched band pass and azimuthal filters for the separation of
950
potential- field anomalies by source region and source type. ASEG Extended Abstracts 2001:
951
15th Geophysical Conference, p. 1–4.
RI PT
949
952
Pushcharovsky, Y. M. 2013. Microcontinents in the Atlantic Ocean. Geotectonics, 47 (4), 241–
954
250.
SC
953
955
Quirk, D. G., Hertle, M., Jeppesen, J. W., Raven, M., Mohriak, W. U., Kann, D. J., Norgaard, M.,
957
Howe, M. J., Hsu, D., Coffey, B., Mendes, M. P. 2013. Rifting, subsidence and continental
958
break-up above a mantle plume in the central South Atlantic. Geological Society, London,
959
Special Publications, 369 (1), 185-214.
960
M AN U
956
Riccommini, C. 2008. Rapid changes of stress-field in the passive continental margin of
962
southeastern Brazil. 33° Int. Geol. Congress, Abstracts, CD-ROM.
EP
963
TE D
961
Sager, W. W., Zhang J., Korenaga J., Sano T., Koppers A.A.P., Mahoney J., Widdowson M.
965
2013. An immense shield volcano within the Shatsky Rise oceanic plateau, northwest Pacific
966
Ocean. Nat. Geosci., 6, p. 976–981.
967
AC C
964
968
Sharp, W. D., Clague, D. A. 2006. 50-Ma initiation of Hawaiian-Emperor bend records major
969
change in Pacific Plate motion. Science, 313 (5791), p.1281–1284.
970
ACCEPTED MANUSCRIPT 44
Souza, K. G., Fontana, R.L., Mascle, J., Macedo, J.M., Mohriak, W.U. & Hinz, K. 1993. The
972
southern Brazilian margin: an example of a South Atlantic volcanic margin. Third International
973
Congress of the Brazilian Geophysical Society, Rio de Janeiro, RJ, 7–11 November 1993.
974
Sociedade Brasileira de Geologia, Rio de Janeiro, 2, p. 1336–1341.
975
RI PT
971
Spector, A. and Grant, F. 1970. Statistical models for interpreting aeromagnetic data. Geophysics,
977
35, p. 293-302.
SC
976
M AN U
978 979
Stica, J.M., Zalán, P.V. and Ferrari, A.L. 2014. The evolution of rifting on the volcanic margin of
980
the Pelotas Basin and the contextualization of the Paraná-Etendeka LIP in the separation of
981
Gondwana in the South Atlantic: Marine and Petroleum Geology, vol. 50, p. 1-21.
982
Syberg, F.R.J. 1972. A Fourier method for the regional-residual problem of potential fields.
984
Geophysical Prospecting, 20, p. 47-75.
TE D
983
EP
985
Stanton, N. S., 2009. Caracterização Crustal da Margem Sudeste Brasileira através de
987
magnetometria e suas implicações tectônicas. Tese de Doutorado, Universidade do Estado do rio
988
de Janeiro, 170 p.
989
AC C
986
990
Tarduno, J. A., Bunge, H.P., Sleep, N., Hansen, U. 2009. The Bent Hawaiian-Emperor Hotspot
991
Track: Inheriting the Mantle Wind. Science, 324, p. 50–53.
992
ACCEPTED MANUSCRIPT 45
Turner, J. P., Wilson, P.G. 2005. Structure and composition of ocean-continent transition at an
994
obliquely divergent transform margin, Gulf of Guinea, West Africa. Petroleum Geoscience, Vol
995
15 2009, p. 305–311.
RI PT
993
996
Unternehr, P., Péron-Pinvidic, G., Manatschal, G., Sutra, E., 2010. Hyper-extended crust in the
998
South Atlantic: in search of a model. Petroleum Geoscience, 16, 207–215.
SC
997
999
Ussami, N., Chaves C.A.M., Marques L.S., Ernesto M. 2013. Origin of the Rio Grande Rise-
1001
Walvis Ridge reviewed integrating palaeogeographic reconstruction, isotope geochemistry and
1002
flexural modelling. Geological Society, London, Special Publications, 369(1), p. 129–146.
M AN U
1000
1003
Williams, S., Müller R.D., Landgrebe, T. C.W., Whittaker, J.M. 2012. An open-source software
1005
environment for visualizing and refining plate tectonic reconstructions using high resolution
1006
geological and geophysical data sets. In: GSA Today, 22, n. 4/5, p. 4-9.
1007
EP
TE D
1004
Wilson, J. T., 1963. Evidence from islands on the spreading of ocean floors. Nature, v. 197, p.
1009
536-538.
AC C
1008
1010 1011
Wilson, J. T., 1965. Submarine Fracture Zones, aseismic ridges and the Interna-tional Council of
1012
Scientific Unions Line: Proposed Western Margin of the East Pacific Ridge. Nature, 207, p. 907–
1013
911.
1014
ACCEPTED MANUSCRIPT 46
1015
Zalán, P. V. and Oliveira, J. A. B., 2005. Origem e evolução estrutural do Sistema de Riftes
1016
Cenozóicos do Sudeste do Brasil. Bol. Geoci. Petrobras, v.13, n. 2, p.269–300.
RI PT
1017
Zalán, P. V., Severino, M. C. G., Rigoti, C. A., Magnavita, L. P., Oliveira, J. A. B., Vianna, A.
1019
R., 2011. An Entirely New 3D-View of the Crustal and Mantle Structure of a South Atlantic
1020
Passive Margin – Santos, Campos and Espírito Santo Basins, Brazil. AAPG Search and
1021
Discovery Article #30177, AAPG Annual Convention and Exhibition 2011, Houston, Texas,
1022
April 10-13, 12 p. (Expanded Abstract).
M AN U
SC
1018
1023
Zhang J., Sager W.W., Korenaga J. 2015. The Shatsky Rise oceanic plateau structure from two-
1025
dimensional multichannel seismic reflection profiles and implications for oceanic plateau
1026
formation. In: Neal CR, Sager WW, Sano T, Erba E (eds) The origin, evolution, and
1027
environmental impact of oceanic large igneous provinces, Geophysical Society of America
1028
Special paper 511, Boulder, p. 103–126.
TE D
1024
EP
1029
Zhang J., Sager W.W., Durkin W. J. 2016. Morphology of Shatsky Rise oceanic plateau from
1031
high resolution bathymetry. Marine Geophysical Research. Advance online publication, in press.
1032 1033 1034
AC C
1030
FIGURE CAPTIONS
1035
Figure 1 - Rio Grande Rise – Walvis Ridge system of aseismic chains in the southern South
1036
Atlantic. The red circles represent the Tristan da Cunha and Gough hotspots.
1037
ACCEPTED MANUSCRIPT 47
Figure 2 – Rio Grande Rise (RGR) and Cruzeiro do Sul Rift (CSR) on the SE margin of South
1039
America. DSDP boreholes: 516F and 21; Seamount Dredging: RC16; WRGR - Western Segment
1040
of the Rio Grande Rise; ERGR – Eastern Segment of the Rio Grande Rise; JCC - Jean Charcot
1041
Seamounts.
RI PT
1038
1042
Figure 3 - Bouguer anomaly map of the Rio Grande Rise (RGR) showing a gravity low in the
1044
region of the aseismic ridge (A). Main gravity domains (GeD1v-GeD6) observed in the RGR
1045
region and the surrounding area (B).
M AN U
SC
1043
1046
Figure 4 - Maps of regional (A) and intermediate (B) Bouguer anomalies of the Rio Grande Rise
1048
(RGR). The dashed black lines on the maps represent the western and eastern segments of the
1049
rise, and the hatched area in A represents the area of interpreted thickened crust under the RGR.
1050
The gray polygon in B represents the abyssal plain between the RGR segments. The red contours
1051
are associated with seamounts and guyots.
EP
1052
TE D
1047
Figure 5 - Residual Bouguer anomaly map (A) showing the locations of the gravity highs
1054
associated with the basement blocks and the flanks of the Cruzeiro do Sul Rift (CSR) (pink
1055
polygons) and the gravity lows associated with areas of thick sedimentary accumulations (blue
1056
polygons) and the aborted rift (white dashed lines). Free-air anomaly map (B) showing gravity
1057
highs (white polygons) associated with the higher Rio Grande Rise (RGR) regions and the flanks
1058
of the aborted rift. The CSR is represented by an extensive negative free-air anomaly (pink
1059
dashed lines). The black dashed line indicates the deep oceanic region dominated by E-W
1060
oriented fracture zones. Profiles A-A' and B-B' are shown in Figure 6.
AC C
1053
ACCEPTED MANUSCRIPT 48
1061 1062
Figure 6 – Bathymetric and gravity profiles A-A' and B-B' across the Rio Grande Rise (RGR).
RI PT
1063
Figure 7 – Overlapping free-air anomalies in the digital terrain model of the Rio Grande Rise
1065
(RGR). The white circle shows the location of the possible volcanic massif.
1066
SC
1064
Figure 8 – Magnetic anomaly reduced to pole (RTP) (A) and RTP anomalies in binary pattern (B)
1068
maps of the Rio Grande Rise (RGR) (dashed lines).
M AN U
1067
1069
Figure 9 –Tilt derivate (TDR) (A) and horizontal gradient amplitude (HGA) (B) maps with
1071
interpreted magnetic lineaments (C) and oceanic fracture zones (D). The continental-oceanic
1072
crust boundary and Isochrone C34 (84 Ma) are represented by white and red lines, respectively.
1073
TE D
1070
Figure 10 – Five evolutionary stages of the Rio Grande Rise (RGR) and the Mid-Atlantic Ridge
1075
from 100 to 55 Ma. Segments 1 to 3 belongs to the WRGR and segments 4 to 7 to the ERGR.
1076
EP
1074
Figure 11 - Temporal-spatial reconstruction of the absolute movements of the South American
1078
and African plates for the periods of 100 Ma (A) and 95 Ma (B). The Mid-Atlantic Ridge is
1079
represented by the black line, and the white circle represents the zone of influence of the Tristan
1080
da Cunha hotspot (radius of 200 km).
AC C
1077
1081 1082
Figure 12 - Temporal-spatial reconstruction of the absolute movements of the South American
1083
and African plates for the periods of 90 Ma (A) and 85 Ma (B). The Mid-Atlantic Ridge is
ACCEPTED MANUSCRIPT 49
1084
represented by the black line, and the white circle represents the zone of influence of the Tristan
1085
da Cunha hotspot (radius of 200 km).
RI PT
1086
Figure 13 - Temporal-spatial reconstruction of the absolute movements of the South American
1088
and African plates for the periods of 80 Ma (A) and 75 Ma (B). The Mid-Atlantic Ridge is
1089
represented by the black line, and the white circle represents the zone of influence of the Tristan
1090
da Cunha hotspot (radius of 200 km).
SC
1087
M AN U
1091
Figure 14 - Temporal-spatial reconstruction of the absolute movements of the South American
1093
and African plates for the periods of 70 Ma (A) and 65 Ma (B). The Mid-Atlantic Ridge is
1094
represented by the black line, and the white circle represents the zone of influence of the Tristan
1095
da Cunha hotspot (radius of 200 km).
1096
TE D
1092
Figure 15 - Temporal-spatial reconstruction of the absolute movements of the South American
1098
and African plates for the periods of 60 Ma (A) and 55 Ma (B). The Mid-Atlantic Ridge is
1099
represented by the black line, and the white circle represents the zone of influence of the Tristan
1100
da Cunha hotspot (radius of 200 km).
AC C
1101
EP
1097
1102
Figure 16 - Free-air gravity anomaly maps showing the tectonic events that controlled the
1103
formation of the Cruzeiro do Sul Rift (CSR) (A) and its northern (NCSR) and southern (SCSR)
1104
segments (B). Areas I and II show major inflections along the CSR and overlaps with the fracture
1105
zones mapped based on the magnetic lineaments.
1106
ACCEPTED MANUSCRIPT 50
Figure 17 – Free-air anomalies map showing the counterclockwise rotations of the compartments
1108
along the Cruzeiro do Sul Lineament in the western and eastern segments of the Rio Grande Rise
1109
(RGR) (white areas I and II). Area III shows a displacement along the rift associated with a
1110
sinistral transcurrent effort in a NE-SW oriented structure, which is well defined on the TDR map
1111
(Figure 9A). The black lines shown in the enlarged regions represent fracture zones.
RI PT
1107
SC
1112
Figure 18 – Free-air anomaly (A) and TDR (B) maps showing the locations of the Jean Charcot
1114
seamount chain (blue - northern segment; red - southern segment) and the FZs, which
1115
compartmentalize this region (black lines). The southern segment (gray area) is shown in detail in
1116
(C).
AC C
EP
TE D
M AN U
1113
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT