Journal of South American Earth Sciences 79 (2017) 431e442

Contents lists available at ScienceDirect

Journal of South American Earth Sciences

journal homepage: www.elsevier.com/locate/jsames

Crustal structure of the Amazonian Craton and adjacent provinces in

Brazil
Diogo Farrapo Albuquerque a, *, George Sand França a, Lucas Paes Moreira b,
Marcelo Assumpça ~o c, Marcelo Bianchi c, Lucas Vieira Barros a,
Cristobal Condori Quispe a, d, Magda Estrela Oliveira a
a
Observato rio Sismolo
gico, Universidade de Brasília (SIS-UnB), Brazil
b
Instituto Federal de Brasília (IFB), Brazil
c
Instituto de Astronomia, Geofísica e Ci^ encias Atmosf
ericas (IAG-USP), Brazil
d
Instituto Geofísico del Perú (IGP), Peru

a r t i c l e i n f o a b s t r a c t

Article history: The study of the crust using receiver functions can provide valuable geological information, such as
Received 13 March 2017 average crustal composition, its formation dynamics and the tectonic evolution of a region, as well as
Received in revised form serve as an initial reference for the generation of seismic wave velocity models to improve earthquake
1 August 2017
location. To fill in gaps in information on the crust of the Amazonian Craton and adjacent provinces in
Accepted 21 August 2017
Available online 9 September 2017
Brazil, we used receiver functions and H-k stacking to estimate crustal thicknesses and the VP/VS ratios.
The results indicate that the crust of the study region is predominantly felsic, with an average VP/VS
around 1.73 and an average thickness of 38.2 km, with a range of 27.4e48.6 km. Minimum curvature
Keywords:
Amazonian Craton
interpolation of the crustal thickness values has made it possible to delimitate of the Amazonian Craton,
Receiver function which corresponds to the area with an average thickness equal to or greater than 39 km. In addition, it
Crustal thickness was possible to identify its potential cratonic blocks, as well as the Paranapanema Block of Parana  Basin.
VP/VS ratio The geometry of the craton, defined by its crustal thickness, is corroborated by the distribution of natural
Brazil seismicity that accompanies its edges. These are related to suture zones between the Amazonian, S~ ao
Francisco/Congo and Paranapanema paleocontinents. The sedimentary basins that have undergone
rifting processes have a thinner crust, usually less than 37 km thick. Due to the great variability of the
results, it was not possible to determine a characteristic value of crustal thickness or VP/VS ratio for each
structural province located in the study region.
© 2017 Elsevier Ltd. All rights reserved.

1. Introduction of stations increased substantially with the deployment of the

The first lithosphere and crustal studies in Brazil using receiver
functions began in 1993 with the project “Seismic studies of con-
tinental lithosphere beneath SE Brazil”, a cooperation between the
University of S~ao Paulo (USP) and the Carnegie Institute of Wash-
ington (James et al., 1993). Since then, many additional projects
have been carried out, but these have been concentrated in the
Northeast, Southeast and part of the Midwest of Brazil, in regions
that historically have had a higher density of seismographic
stations.
This situation has been changing since 2003, when the number
* Corresponding author.
E-mail address: diogofarrapo@gmail.com (D.F. Albuquerque).
Petrobras supported Brazilian Seismic Network (RSBR). Currently,
the RSBR has more than 90 stations and approximately 22 are
located in the North and Center-West regions (Fig. 1), which have
real-time seismographic monitoring.
The motivation of this work emerged from the need to fill in
gaps of crustal thickness and VP/VS ratio values of the Amazonian
Craton and adjacent provinces. These parameters can provide
important information on average crustal composition, formation
dynamics, and tectonic evolution of a region and also serve as an
initial reference for the generation of seismic wave propagation
velocity models to improve earthquake location.
We have used the receiver function (RF) (Ligorría and Ammon,
1999) and H-k stacking (Zhu and Kanamori, 2000) to estimate the
crustal thickness and VP/VS ratio of the Amazonian Craton and

http://dx.doi.org/10.1016/j.jsames.2017.08.019
0895-9811/© 2017 Elsevier Ltd. All rights reserved.
432 D.F. Albuquerque et al. / Journal of South American Earth Sciences 79 (2017) 431e442

Fig. 1. Brazilian structural provinces (Almeida et al., 2000) and the location of seismographic stations used in this study. The inset map in the lower left corner highlights hypotheses
found in prior literature for the region and shows the location of the data points. Caption acronyms: Amazonas Province (AmPr), Amazonian Craton (AmCr), Parecis Basin (PrBs),
Tocantins Province (ToPr), Sa ~o Francisco Craton (SFCr), Parnaíba Basin (PbBs) and Parana Basin (PnBs). Networks: Brazilian Seismographic Network (RSBR); Seismological Obser-
vatory at the University of Brasília (SIS-UnB); Tectonic Studies of the Pantanal, Chaco and Parana  Basin (PCPB-USP); Tectonic Studies of Parecis Basin (ETBP-CNPq).

other geological provinces using data from 49 stations belonging to
4 networks (Fig. 1): 22 stations of the Brazilian Seismic Network
(RSBR); 9 stations of the Seismological Observatory, University of
Brasília (SIS-UnB); 6 stations of the Tectonic Studies of Parecis Basin
(ETBP); 12 stations of the Structural Studies Network of the Pan-
tanal, Chaco and Parana  Basin (PCPB-USP).
França (2003) presented the first crustal thickness (H)
compilation map of the Southeast and part of the Center-West of
Brazil. Building on this, Pava~o et al. (2012) compiled H and VP/VS
values obtained by RFs and built a contour map for Brazil.
Assumpça ~o et al. (2013) updated this work by including H estimates
for all the South America from deep seismic refraction (Berrocal
et al., 2004; Soares et al., 2006), RF (Niu et al., 2007; Bianchi,
2008; Lloyd et al., 2010; Albuquerque et al., 2011; Rosa et al.,
 D.F. Albuquerque et al. / Journal of South American Earth Sciences 79 (2017) 431e442
2012) and velocity models generated by joint inversion of receiver
function and surface wave dispersion (An and Assumpça ~o, 2006;
 et al., 2008; Lima, 2011).
Julia
To update these previous compilations, the current study first
investigated and compiled data from published papers (Trabant
et al., 2012; Trindade et al., 2014), conference proceedings (Costa
and França, 2008; Assumpça ~o et al., 2015), and unpublished
monographs (Nascimento, 2011; Albuquerque, 2014). This new
compilation has a total of 100 crustal thickness values and 59 VP/VS
estimates (Fig. 1). The difference between the number values for
each parameter is due to the lack of VP/VS information in many
papers.
In Fig. 1, the inset map shows that most of the points in this
compilation are concentrated on the edges of the Amazonian
Craton, leaving a wide region without coverage, highlighting the
need for more coverage which motivates the present study. The
stations used in this study are mostly located where there is no
prior information of crustal thickness and/or VP/VS.
Fig. 2 presents the variation of H and VP/VS for each province and
the statistics of those parameters. The graph in 2a points out that it
is not possible to establish a characteristic value for the H and VP/VS
parameters (Fig. 2a), since there is a large range in the estimates
and averages. In graphs 2b and 2c, it is evident that the thicknesses
are concentrated between 35 and 47 km, while the VP/VS values are
concentrated between 1.66 and 1.76.
The average crustal thickness of the region is 41.6 km, ranging
from 30.6 km to 53.0 km, with the minimum and maximum both in
Tocantins Province. This value is near to 41.5 km estimated by
Fig. 2. a) Average and absolute variation in the values of crustal thickness (H) and VP/VS ratio. b) Frequency distribution (%) of crustal thickness. c) Frequency distribution of VP/VS.
433
Christensen and Mooney (1995) for shields and platforms. The
average VP/VS estimate is 1.75, ranging from 1.54 (Amazonas
Province) to 1.90 (Parecis Basin). The average VP/VS can be attrib-
uted to a generally felsic crustal composition, but the maximum
and minimum indicate that the composition ranges from felsic to
mafic (Musacchio et al., 1997).
Hypotheses have been raised in seismological studies of the
Amazonian Craton to explain some of these observations. Krüger
et al. (2002), using a local seismographic network (Fig. 1, circle
number 1), found an abrupt crustal thickness variation between the
northern portion of the Amazonian Craton (Guianas Shield) and the
Amazonas Basin, which may indicate that Moho dips approxi-
mately 20
towards the north.
Lloyd et al. (2010) found a positive correlation between the age
of some regions of the craton and its thickness, indicating that
Moho is deeper in places where the crust dates from 2 to 3 Gyr. Rosa
et al. (2016) found that the crustal thickness of the Amazonian
Craton decreases in the NE-SW direction (Fig. 1, circle number 2)
and that its northern portion (Guianas Shield) has a remnant root
resulting from its tectonic history.
Assumpça ~o et al. (2013) found some evidence of a crustal
thinning in the region known as the Paraguai-Araguaia Fold Belt
(Fig. 1, circle number 3), which can represent a remnant feature of a
suture zone between the Amazon and Atlantic crustal blocks.
Agurto-Detzel et al. (2015) pointed out that the occurrence of
earthquakes in Brazil is greater in regions with a crust thinner than
35 km, and smaller in those with thickness ranging from 35 to
45 km.
434 D.F. Albuquerque et al. / Journal of South American Earth Sciences 79 (2017) 431e442
2. Geological settings
Brazil is located in the interior of the South American Platform,
defined by Almeida et al. (2000) as the continental part of the crust
that has remained stable and relatively undeformed during the
latest orogenic event, which occurred in Mesozoic to Cenozoic eras
(Caribbean and Andes orogeny).
The South American platform originated from the arrangement
of Archean and Proterozoic crustal blocks during three main
orogenic events: (1) Transamazonian (Paleoproterozoic e 2.2 to 1.8
Gyr), (2) Late Mesoproterozoic (1.3e0.95 Gyr) and (3) Brasiliano/
Pan-African (0.9e0.5 Gyr). The latter is considered the main
event that gave birth to the present global pattern of tectonic
components (cratonic nuclei and folding belts) and the formation of
the structural framework of the platform when was still part of the
super continent Gondwana (Almeida et al., 2000).
In the context of structural provinces (Almeida et al., 2000), the
geology of the North and Center-West regions of Brazil (Fig. 1) is
dominated by the Amazonian Craton, one of the largest and least
studied Precambrian regions in the world, with an extension of
5,600,000 km2, of which 4,000,000 km2 are in Brazilian territory.
According to Pimentel and Fuck (1992), the Amazonian Craton in
Brazil is limited to the East (Baixo Araguaia Groups), South and
Southeast (Alto Paraguai, Cuiaba  and Corumba  Groups) by rocks
generated during the Brasiliano Orogenic Cycle (930e540 Myr).
Two different models have been used to compartmentalize the
craton: one geophysical-structural and one geochronological. The
first considers the craton to be the union of twelve Archean blocks,
with its margins delimited by nineteen collisional belts as old as
those blocks (Hasui et al., 1984). The second considers geochrono-
logical aspects based in U-Pb, Sm-Nd and Rb-Sr dating (Amaral,
1974; Cordani et al., 1979).
Based on the geochronological model, Santos (2003) divides the
Amazonian Craton in seven provinces (Fig. 3): Caraja s (3.10e2.53
Gyr), Amazo ^nia Central (2.60e1.70 Gyr), Transamazonas
(2.25e2.00 Gyr), Tapajo  s-Parima (2.10e1.87 Gyr), Rio Negro
(1.86e1.52 Gyr), Rondo ^nia-Juruena (1,81 to 1,52 Gyr), Sunsas (1,45
to 0,99 Gyr) and the Phanerozoic covers.
The Amazonas and Tocantins Provinces and the Parana , Parecis
and Parnaíba basins complete the geology of the study region.
Other geological aspects related to these structural provinces will
be discussed with the results below.
3. Methods
The receiver function method is used to isolate the relative
response of the Earth structure near the receiver (the seismic
sensor), using three-component seismograms of distant earth-
quakes, located between the distances of 30
and 90
(Ammon,
1991). It is based on the principle that when a P wave train rea-
ches the base of the crust (Moho) it crosses from a medium with
higher velocity to a slower one, causing part of its energy to be
converted into a S wave (phase Ps), generating additional phases
(PpPms and PpSms þ PsPms). The resulting waveform is a time
series with P phases converted to S (Ps) and by reverberations
(Langston, 1979).
The RF is essentially a deconvolution and there are several
techniques to perform this deconvolution, each with its particu-
larities. In this case, we have applied the iterative time-domain
deconvolution (ITERDECON) developed by Ligorría and Ammon
(1999).
The ITERDECON technique is based on the least-squares mini-
mization of the difference between the observed horizontal seis-
mogram and a predicted signal generated by the convolution of an
iteratively updated spike train with the vertical-component
seismogram (Ligorría and Ammon, 1999). To improve the signal/
noise ratio, the technique requires a Gaussian filter factor, which
controls the bandwidth of the signal, and a preset number of iter-
ations to perform. In this case, we applied a Gaussian factor equal to
2.5 and 200 iterations.
When compared to other techniques, such as the spectral divi-
sion deconvolution (Langston, 1979) approach, the main advantage
of ITERDECON is the generation of RFs without the need to establish
complex water level values (Clayton and Wiggins, 1976), which can
vary according to the signal-to-noise ratio of each seismogram. This
prevents, for example, the additional step of visual selection of the
waveforms to test the optimal water level. In addition, it is possible
to automatically eliminate bad RFs based on the percentage fit of
the observed power in the horizontal seismograms calculated by
the program.
To estimate H and VP/VS ratio from the RFs, we used the H-k
stacking method (Zhu and Kanamori, 2000). In H-k, the radial RFs
are stacked on the predicted arrival time of Ps and multiples
(PpPms and PpSms þ PsPms) using different H and VP/VS values.
The best result is reached when all the three phases are stacked
coherently (maximum in the H-k function).
As an initial parameter to run H-k stacking, we used a P velocity
(VP) equal to 6.4 km/s, which is the average for Brazil (França, 2003;
Berrocal et al., 2004; Soares et al., 2006; Pava ~o et al., 2012;
Assumpça ~o et al., 2013). As recommended by Zhu and Kanamori
(2000), we also used the following weighting factors: w1 ¼ 0.7,
w2 ¼ 0.2 and w3 ¼ 0.1, corresponding to the phases Ps, PpPms and
PpSms þ PsPms, respectively. That choice is also based on the
amplitude of Ps and multiples.
H-k stacking was not used on data from stations of PCPB-USP
(Fig. 1) because the dataset was too small. Instead, we applied a
move-out correction in the RFs for a distance of 66
and a slowness
of 6.4 s/degree and then we stacked all the waveforms.
4. Results
Table 1 presents the crustal thicknesses and VP/VS ratios esti-
mated using radial RFs (Ligorría and Ammon, 1999) and the H-k
stacking (Zhu and Kanamori, 2000) for each structural province in
the North and Center-West of Brazil.
The crustal thickness in the North and Central-West regions of
Brazil is quite variable, with the lowest value found for the Ama-
zonas Province (27.4 km) and the largest for the Amazonian Craton
(48.6 km). The VP/VS ratio reaches the lowest value for the Ama-
zonas Province (1.41) and the largest for the Parecis Basin and
Amazonas Province (1.86).
Figs. 4 and 5 show the radial receiver functions and H-k stacking
estimates for the stations ITTB, PTLB and VILB, which are located in
the Amazonian Craton or in its basins. These stations were the ones
with the smallest associated error in H and VP/VS results.
The VP/VS results indicate that the crust in the craton is quite
variable in terms of composition, ranging from felsic (lower than
1.73) to mafic (greater than 1.80), according to the relation stated by
Musacchio et al. (1997).
Taking into account the large range in the estimates for both
parameters, it is not possible to establish a characteristic value for
each structural province. Therefore, they are not directly related to
the geological information that has been used to define their bor-
ders, although they are important for describing the crustal and
lithospheric evolution related to various continental arrangements
in the past.
It can be seen from Table 1 that most crustal thickness estimates
have errors below 3.0 km. The smallest error is associated with the
PTLB station (H ¼ 41.1 ± 0.3 km) and the largest error with PAR3
(H ¼ 43.0 ± 4.6 km).
 D.F. Albuquerque et al. / Journal of South American Earth Sciences 79 (2017) 431e442 435

Fig. 3. Amazonian Craton and its geochronological provinces (Santos, 2003).

When we consider the region that includes the Amazonian
Craton, and its basins (Amazonas Province and Parecis Basin), the
average crustal thickness is 38.2 km, ranging from a minimum of
27.4 to a maximum of 48.6 km. The average H is lower than the
estimated by Christensen and Mooney (1995) for shields and
platforms.
4.1. Variation of crustal thickness and geological implications
From the results obtained in the literature and in the present
work, we have generated three crustal thickness interpolation
maps (Fig. 6), using the minimum curvature method (Smith and
Wessel, 1990).
In order to generate a Moho surface, the interpolation method
implemented in Generic Mapping Tools (Wessel et al., 2013) re-
quires two input parameters: grid spacing or increment (I), and
tension factor (T), where 0  T  1. The first defines the spatial
resolution of the variable to be interpolated and the second, the
smoothing of the maximum and minimum of the interpolation
surface.
We chose the increment based on the lateral resolution of the
RF, which corresponds approximately to the depth of the
436 D.F. Albuquerque et al. / Journal of South American Earth Sciences 79 (2017) 431e442

Table 1
Crustal thickness (H) and VP/VS estimates for each structural province.

Structural province Station Network Latitude (
) Longitude (
) Elevation (m) H (km) VP/VS Average H (km) Average Vp/Vs

Amazonian Craton NPGB RSBR 07.05 55.36 000 34.4 ± 0.9 1.73 ± 0.02 41.0 1.71
PRPB RSBR 06.17 49.82 265 37.7 ± 0.4 1.73 ± 0.02
PTLB RSBR 15.45 59.14 072 41.1 ± 0.3 1.68 ± 0.01
BOAV RSBR 02.40 60.52 114 42.8 ± 1.6 1.72 ± 0.07
SGCB RSBR 00.12 67.03 070 41.3 ± 3.0 1.68 ± 0.06
VBST PCPB-USP 14.53 60.02 235 48.6* 1.73

Amazonas Province CZSB RSBR 07.73 72.70 196 27.4 ± 4.5 1.41 ± 0.15 37.9 1.71
ETMB RSBR 09.82 66.21 118 44.2 ± 2.5 1.71 ± 0.10
IPMB RSBR 17.98 48.21 706 37.8 ± 1.0 1.81 ± 0.02
ITTB RSBR 04.37 55.73 118 42.7 ± 0.6 1.73 ± 0.01
MACA RSBR 03.16 60.68 075 42.0 ± 1.8 1.86 ± 0.12
MALB RSBR 01.85 54.26 027 40.2 ± 0.8 1.82 ± 0.03
MCPB RSBR 00.36 52.06 127 33.4 ± 1.1 1.82 ± 0.03
TBTG RSBR 04.19 69.91 091 38.2 ± 2.5 1.57 ± 0.04
TEFE RSBR 03.51 64.63 052 35.1 ± 2.7 1.68 ± 0.09

Parecis Basin PDRB RSBR 11.61 56.73 330 40.4 ± 2.2 1.48 ± 0.03 37.2 1.75
VILB RSBR 12.95 60.20 434 40.4 ± 0.5 1.78 ± 0.01
PAR1 ETPB-CNPq 11.55 57.87 314 33.4 ± 2.4 1.79 ± 0.10
PAR2 ETPB-CNPq 12.10 58.33 374 38.2 ± 3.6 1.75 ± 0.10
PAR3 ETPB-CNPq 11.47 58.72 347 43.0 ± 4.6 1.78 ± 0.09
PAR4 ETPB-CNPq 11.53 59.06 352 e e
PAR5 ETPB-CNPq 11.58 57.39 255 30.0 ± 4.5 1.80 ± 0.11
PAR6 ETPB-CNPq 11.52 58.17 260 34.9 ± 2.7 1.86 ± 0.13

Tocantins Province ARAG RSBR 15.71 51.81 237 32.2 ± 0.4 1.74 ± 0.02 38.9 1.72
PEXB RSBR 12.11 48.30 350 39.6 ± 0.8 1.69 ± 0.02
SALV RSBR 15.90 55.69 213 35.5 ± 2.1 1.66 ± 0.07
SNDB RSBR 11.97 51.29 252 37.4 ± 1.3 1.71 ± 0.03
SFA1 SIS-UnB 17.97 47.70 829 43.2 ± 2.0 1.70 ± 0.04
CAN1 SIS-UnB 13.42 48.09 315 43.1 ± 2.8 1.71 ± 0.07
CAN3 SIS-UnB 13.08 48.29 430 40.5 ± 2.3 1.67 ± 0.04
SSV2 SIS-UnB 12.85 48.21 315 42.6 ± 1.8 1.77 ± 0.04
MR07 SIS-UnB 13.77 49.16 376 39.0 ± 1.9 1.71 ± 0.04
NBRS PCPB-USP 14.32 55.78 490 41.5* 1.73
PTET PCPB-USP 15.38 57.16 185 39.0* 1.73
POCN PCPB-USP 16.62 56.73 116 29.3* 1.73
PANT PCPB-USP 18.99 56.62 271 35.6* 1.73
BDQN PCPB-USP 20.45 56.75 180 41.9* 1.73
MURT PCPB-USP 21.66 57.61 123 43.6* 1.73

Parnaíba Basin TMAB RSBR 02.37 48.10 026 38.6 ± 2.8 1.73 ± 0.07 37.6 1.73
SMTB RSBR 08.86 47.59 292 36.9 ± 0.4 1.84 ± 0.01
PAL1 SIS-UnB 06.64 47.56 320 31.4 ± 2.2 1.64 ± 0.06
PAL2 SIS-UnB 06.75 47.33 300 35.4 ± 1.5 1.72 ± 0.04
LAJE SIS-UnB 09.98 48.32 294 41.7 ± 3.6 1.72 ± 0.07
TUC4 SIS-UnB 04.25 49.54 100 41.8 ± 2.0 1.75 ± 0.08

 Basin
Parana DVLD PCPB-USP 16.64 52.16 548 38.6 1.73 36.4 1.73
RVDE PCPB-USP 19.03 54.94 470 29.8 1.73
RPRD PCPB-USP 20.48 53.70 343 39.0 1.73
ANTJ PCPB-USP 22.00 56.00 532 33.1 1.73
AMBA PCPB-USP 22.93 54.99 330 41.5 1.73

discontinuity to be identified (Ammon et al., 1990). As the Moho
discontinuity is what interests us for estimating the crustal thick-
ness, the increment was defined as the average of the values of H
(41 km), obtained from the references and the new results (Table 1).
Tests were then performed to define T, varying its value from
0 to 1. The interpolation results showed that 0  T  0.25 is
acceptable and does not considerably change the resulting surface.
As the crustal thickness variation, in general, has a smoothed
behavior (in other words it works as a smoothing constraint) the T
parameter was defined to be zero. Fig. 6 presents the interpolation
of crustal thickness using estimates from the references (a), those
obtained in this work (b), and from the combination of both (c).
There are some fundamental differences between the maps a)
and b) in Fig. 6. It is quite clear that in 6a, the stations are located at
the borders of the Amazonian Craton, leaving a wide region without
any information, which generates a smoothed Moho surface. The
map in 6b has better sampling in the Amazonian Craton, indicating
areas with thicker crust (blue areas).
Although there are points of crustal thinning within the
Amazonian Craton, it is possible to define the craton as the
northwestern portion of the black dashed line (Fig. 6c), which has a
thickness equal to or greater than 39 km. These borders are in
accordance with those defined by Rosa et al. (2016) and Fuck et al.
(2008) from geological and geophysical data.
Rosa et al. (2016)’s hypothesis about the crustal thinning in the
Amazonian Craton along a NE-SW direction (Fig. 1, circle number
2), cannot be verified in the AB profile (Fig. 7). A central domain can
be established in the Amazonas Province, with a crustal thickness
ranging from 40 to 45 km, composed of the Solimo ~es and Ama-
zonas basins. In addition, the definition of basin boundaries by
crustal thickness variation also coincides with the approximate
location of the arches that separate them, such as the Purus Arch,
between the Amazonas and Solimo ~ es basins, and the Iquitos Arch,
considered to be the eastern boundary of the Acre Basin, which has
 D.F. Albuquerque et al. / Journal of South American Earth Sciences 79 (2017) 431e442 437

Fig. 4. Radial receiver functions ordered by back azimuth for the stations ITTB, PTLB and VILB. The darkened regions indicate the arrival interval of the phases of Ps and its multiples.

Fig. 5. H-k stacking applied to the receiver functions for the stations ITTB, PTLB and VILB. The red ellipse indicates the crustal thickness (H) and VP/VS estimate. (For interpretation of
the references to colour in this figure legend, the reader is referred to the web version of this article.)

a relatively thin crust, around 27 km, directly related to the flexural
subsidence in the South American Plate caused by the Andean
chain.
The regions of greater crustal thickness (H  43 km), marked by
numbers 1, 3, 4, 5, 6 and 9 in Fig. 6, indicate the possible blocks that
formed the Amazonian Craton. The number 14 indicates the Para-
napanema block, located in the Parana  Basin, formed approxi-
mately between 1.0 Gyr and 850 Myr (Mantovani and Quintas,
1996). Geophysical evidence obtained from gravimetric surveys in
the Paran a Basin (Mantovani et al., 2005) also corroborate its
existence.
Matos and Brown (1992) carried out a seismic reflection survey
near the area marked by circle number 3 (Fig. 6c) and they esti-
mated a crustal thickness of roughly 45 km, consistent to the value
in the H interpolation map (Fig. 6c) and in the AB profile (Fig. 7).
Daly et al. (2014) carried out a deep seismic reflection profile
438 D.F. Albuquerque et al. / Journal of South American Earth Sciences 79 (2017) 431e442

Fig. 6. Minimum curvature interpolation map using crustal thickness estimates in the literature (a), obtained in this work (b) and the combination of both (c). The numbers point
out features discussed in the text. The lines AB and CD represent the profiles in Fig. 7. Caption acronyms: Amazonas Province (AmPr), Amazonian Craton (AmCr), Parecis Basin (PrBs),
Tocantins Province (ToPr), Sa~o Francisco Craton (SFCr), Parnaíba Basin (PbBs) and Parana  Basin (PnBs).

Fig. 7. Topographic and crustal thickness variation along AB and CD profiles. The dimensions were exaggerated to enhance features.

from west to east, beginning at the eastern boundary of the
Amazonian Craton, crossing the northern portion of the Tocantins
Province and the Parnaíba Basin, and reaching the western limit of
the Borborema Province. They estimated a crustal thickness of
around 40 km for the Amazonian Craton, reaching 45 km in the
Parnaíba Basin. This observation is in accordance with the
40e45 km variation indicated by the isolines in Fig. 6c (near circle
number 9).
In general, regions of thinner crust are located in sedimentary
basins (Fig. 6c, numbers 2, 7, 11, 13 and 17), indicating that some
tectonic process of crustal extension may be related to their origin
or evolution.
Comparing topography with crustal thickness variation (Fig. 7,
profiles AB and CD), it is clear that we cannot correlate them. This
means that the region is not in isostatic compensation.
In both the interpolation map (Fig. 6c) and the CD profile (Fig. 7)
we see an increase in crustal thickness in the direction from the
Amazonas Province (AmPr) to the Amazonian Craton (AmCr), in the
area known as the Guianas Shield (GS). This observation corrobo-
rates the hypotheses of Krüger et al. (2002) and Rosa et al. (2016),
who suggest, respectively, an increase in crustal thickness in the
AmPr-GS direction, with a sloping Moho and a remnant root from
the tectonic events that formed the craton. Niu et al. (2007) esti-
mated a thickness of 36.5 km below a station located more than
400 km from BOAV, which means that the tendency for an
increasing H is not observed further north of that station.
It was not possible to correlate the geochronology of the
Amazonian Craton provinces (Fig. 8) and their crustal thicknesses,
 D.F. Albuquerque et al. / Journal of South American Earth Sciences 79 (2017) 431e442 439

Fig. 8. Crustal thickness interpolation map combining all the results from the Amazonian Craton and its geochronological provinces. The caption acronyms represent the
geochronological provinces (Santos, 2003) of the craton: Carajas (Crjz), Amazo
^nia Central (AmCt), Transamazonas (TrAm), Tapajo
s-Parima (TaPa), Rio Negro (RiNg), Rondo
^ nia-
Juruena (RoJu) e Sunsas (Suns).

as noted by Lloyd et al. (2010), because there are regions of thicker complex inversion parameters and because it is better sampled,
crust in younger provinces, such as Rondo ^nia-Juruena, and of especially in the Amazonian Craton.
thinner crust in older provinces, such as Transamazonas. Even
when we consider the same province, such as Tapajo s-Parima,
4.2. Seismicity and crustal thickness
there are variations large enough to falsify this correlation.
The values of H in the Parecis Basin (Fig. 8) are close to those
Brazil is located in the stable portion of the South American
reported by Bahia et al. (2007) and Faria (2015), which found a
Plate and because of this, it has low seismicity compared to regions
variation between 30 and 40 km. Additionally, there is a tendency
on plate borders and even compared to similar tectonic regions of
for crustal thinning from the borders to the center of the basin,
plate interiors, such as the central and eastern United States,
which may be related to crustal or lithospheric thinning occurring
Australia, India, and China (Assumpça ~o et al., 2014).
as the asthenosphere rises as a consequence of the rifting process
The causes of intraplate seismicity are still a matter of discus-
that occurred in the Parecis Basin (Bahia et al., 2007).
sion, since it is often not possible to associate this seismicity with
Fig. 9a shows the Bouguer Anomaly map of Brazil extracted from
identifiable geological structures. However, this seismicity tends to
WGM2012 Earth's gravity anomalies (Balmino et al., 2012).
concentrate around craton boundaries (Mooney et al., 2012) and in
Comparing Figs. 6c and 9a, we can see that regions with a high
regions of accumulation of stress in the upper crust due to litho-
Bouguer anomaly do not usually have the thin crust we expect for
spheric thinning, as in the Central Brazil region (Assumpça ~o et al.,
the case of isostatic compensation.
2004, 2014).
Comparing the crustal thickness interpolation map in Fig. 6c
In Fig. 10, the map shows the distribution of earthquakes
and the crustal thickness model GEMMA 1.0 (Fig. 9b), derived from
occurring in Brazil between 1950 and 2016, with magnitudes equal
gravity satellite data (Reguzzoni and Sampietro, 2015), it is
or greater than 3.0. All the information was taken from the Brazilian
possible to see that GEMMA 1.0 estimates a thinner crust
Seismic Catalog (SISBRA), provided by the Seismological Observa-
(H < 37 km) for the Amazonian craton and its basins. This in-
tory at the University of Brasília (SIS-UnB).
dicates that GEMMA 1.0 is unrealistic and needs adjustments to its
Although the distribution of recorded seismicity is conditioned
inversion constraints. Therefore, the map in Fig. 6c represents a
by the arrangement of seismographic stations along the Amazonian
more reliable crustal model both because it does not rely on
Craton, which is quite sparse and irregular, it is possible to see that
440 D.F. Albuquerque et al. / Journal of South American Earth Sciences 79 (2017) 431e442
Fig. 9. a) Bouguer anomaly map for the study region extracted from WGM2012 Earth's gravity anomalies (Balmino et al., 2012). b) Minimum curvature interpolation map using
GEMMA 1.0 values (Reguzzoni and Sampietro, 2015). The Caption acronyms represent the geochronological provinces (Santos, 2003) of the craton: Caraja
(AmCt), Transamazonas (TrAm), Tapajo s-Parima (TaPa), Rio Negro (RiNg), Rondo
^ nia-Juruena (RoJu) e Sunsas (Suns).
seismicity is concentrated along areas of generally thinner crust
(H < 39 km), which accompany the Amazonian Craton borders
(Fig. 10, darkened region). Thus, its occurrence is correlated with
the limits or suture zones between the craton and other structural
provinces, which corroborates the thesis of Mooney et al. (2012)
and Assumpça ~o et al. (2004, 2014).
It is also possible to find that there is a concentration of earth-
quakes where there are abrupt variations in the thickness of the
crust, as between the Parecis Basin and the Amazonian Craton as
well as between this and the Amazonas Province. However, there is
still insufficient evidence to state that these variations in thickness
have a direct influence on the observed seismicity.
With the exception of the thinner crust located in the Acre Basin,
where deep earthquakes are related to the subduction of the Nazca
Plate, the seismicity observed in Brazil has a number of different
causes, indicating that there is not a strong correlation between
areas with crust that has thicknesses less than 35 km and the
occurrence of earthquakes, as cited by (Agurto-Detzel et al., 2015).
In fact, this trend is only observed for regions with crustal thick-
nesses equal to or less than 39 km, but located on the borders of the
Amazonian Craton.
5. Discussions and conclusions
In general, receiver functions (Ligorría and Ammon, 1999),
combined with the Hk stacking (Zhu and Kanamori, 2000), pro-
vided crustal thickness and VP/VS estimates with good precision.
The new results fill gaps in crustal information, mainly for the
Amazonian Craton, and have made possible the creation of a more
accurate crustal thickness model for whole study region.
The average crustal thickness of the Amazonian Craton and
adjacent provinces was estimated at 38.2 km and, therefore, in
accordance with the global average for continental crust (39.2 km),
and for shields and platforms (41.5 km) according to Christensen
and Mooney (1995). In addition, the average VP/VS ratio, esti-
mated at 1.73, suggests a dominantly felsic crustal composition, but
locally ranging from felsic to mafic following the classification of
Musacchio et al. (1997).
s (Crjz), Amazo
^ nia Central
Sedimentary basins that have undergone some the process of
crustal rifting during their formation or tectonic evolution, such as
the Parecis and Parnaíba basins, have a crust with average thick-
nesses below 37 km. However, more data is necessary to corrobo-
rate this finding, because these regions have a smaller number of
stations.
Although it is possible to differentiate between cratonic regions
and sedimentary basins, due to the great variability in crustal
thickness and VP/VS ratio, it is not possible to define a characteristic
value for each parameter when we consider the context of the
structural provinces.
The regions with thicker crust, often over 43 km, located in the
Amazonian Craton and Paran a Basin, may be related to ancient
cratonic blocks, which were part of the Amazon, Sa ~o Francisco/
Congo and Paranapanema paleocontinents. The identification of
the Paranapanema Block is also corroborated by geophysical data
(Mantovani et al., 2005).
The hypothesis by Rosa et al. (2016) calling for crustal thinning
in the Amazonian Craton along a NE-SW direction is not supported
by the new data. Most of the crustal thickness results are in
accordance with seismic refraction surveys carried out in the
Amazonian Craton (Matos and Brown, 1992; Daly et al., 2014) and
the adjacent provinces (Berrocal et al., 2004; Soares et al., 2006).
It was not possible to correlate the geochronology of the
Amazonian Craton provinces (Fig. 8) and their crustal thicknesses,
as noted by Lloyd et al. (2010), because there are regions of thicker
crust in younger provinces, such as Rondo ^ nia-Juruena, and of
thinner crust in older provinces, such as Transamazonas.
In terms of Bouguer Anomaly, it was not possible to correlate
gravity anomalies with crustal thickness. Comparing our results
with a crustal model based on the inversion of gravity satellite data
(Reguzzoni and Sampietro, 2015), we find that our model (Fig. 6c) is
more accurate.
The limits of the Amazonian Craton, in terms of crustal thick-
ness, follow the patterns in seismicity distribution at its edges.
These patterns relate to suture zones between the paleocontinents
that collided in the Brasiliano event.
There is no correlation indicating that regions with crustal
 D.F. Albuquerque et al. / Journal of South American Earth Sciences 79 (2017) 431e442 441

Fig. 10. Interpolation of the data obtained in this work and from the literature. The red circles represent earthquakes which occurred between 1950 and 2016. The dashed black line
indicates the crustal boundaries of the Amazonian Craton and the darkened region indicates the suture zones between the craton and the other provinces. The acronyms represent
the Brazilian structural provinces: AmPr (Amazonas Province), AmCr (Amazonian Craton), PbBs (Parnaíba Basin), PnBs (Parana Basin), PrBs (Parecis Basin) and ToPr (Tocantins
Province). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

thickness equal to or less than 35 km have higher seismicity, as
suggested by Agurto-Detzel et al. (2015) or the opposite for crust
thicker than 35 km.
Acknowledgements
DFA thanks CAPES for supporting this research (G.N. 1477700).
We thank the Seismological Observatory at the University of Bra-
sília, and the Brazilian Seismographic Network (RSBR) for providing
the seismological data. We also thank all the station operators,
especially Paulo Araújo de Azevedo, and those who provided
computing support, especially Fernando Paiva. GSF thanks for his
PQ grant (307251/2016-0) and this research received support by
MCTI/CNPq-Universal “Tectonics studies in the Parecis Basin”
456560/2014-9.
References
Agurto-Detzel, H., Assumpç~
ao, M., Bianchi, M., Pirchiner, M., 2015. Intraplate
seismicity in mid-plate South America: correlations with geophysical litho-
spheric parameters. Geol. Soc. Lond. Spec. Publ. 432, 73e90. http://dx.doi.org/
10.1144/SP432.5.
Albuquerque, D.F., 2014. Determinaça ~o de espessuras crustais utilizando o me todo
de funça~o do receptor. Universidade de Brasília, Brasília.
Albuquerque, D.F., Pava~o, C.G., Silveira, R.T.G. da, Santos, I.G. dos, França, G.S., 2011.
Crustal thickness estimatives and vp/vs ratio using receiver functions. In:
Chaves, A.C. (Ed.), 12th International Congress of the Brazilian Geophysical
Society & EXPOGEF. Society of Exploration Geophysicists and Brazilian
Geophysical Society, Rio de Janeiro, pp. 2018e2021. http://dx.doi.org/10.1190/
sbgf2011-417.
Almeida, F.F.M. de, Brito Neves, B.B. de, Dal Re  Carneiro, C., 2000. The origin and
evolution of the South American platform. Earth-Science Rev. 50, 77e111.
http://dx.doi.org/10.1016/S0012-8252(99)00072-0.
Amaral, G., 1974. Geologia pre-cambriana da regi~ ao amazo ^ nica. Universidade de S~ao
Paulo.
Ammon, C.J., 1991. The isolation of receiver effects from teleseismic P waveforms.
Bull. Seismol. Soc. Am. 81, 2504e2510.
Ammon, C.J., Randall, G.E., Zandt, G., 1990. On the nonuniqueness of receiver
function inversions. J. Geophys. Res. 95, 15303. http://dx.doi.org/10.1029/
JB095iB10p15303.
An, M., Assumpça ~o, M., 2006. Crustal and upper mantle structure in the intra-
cratonic Parana Basin, SE Brazil, from surface wave dispersion using genetic
algorithms. J. South Am. Earth Sci. 21, 173e184. http://dx.doi.org/10.1016/
442 D.F. Albuquerque et al. / Journal of South American Earth Sciences 79 (2017) 431e442

j.jsames.2006.03.001. Langston, C.A., 1979. Structure under Mount Rainier, Washington, inferred from
Assumpç~ ao, M., Bianchi, M., Albuquerque, D.F., França, G.S.L., 2015. Crustal thickness teleseismic body waves. J. Geophys. Res. 84, 4749. http://dx.doi.org/10.1029/
map in South America: an updated version. In: 1st Brazilian Symposium on JB084iB09p04749.
Seismology. Brasília, pp. 1e4. Ligorría, J.P., Ammon, C.J., 1999. Iterative deconvolution and receiver function
Assumpç~ ao, M., Bianchi, M., Julia , J., Dias, F.L., Sand França, G., Nascimento, R., estimation. Bull. Seismol. Soc. Am. 89, 1395e1400.
Drouet, S., Pava ~o, C.G., Albuquerque, D.F., Lopes, A.E.V., 2013. Crustal thickness Lima, M.V.A.G. de, 2011. Emprego do me todo de refraç~ ao sísmica profunda para a
map of Brazil: data compilation and main features. J. South Am. Earth Sci. 43, determinaç~ ao da estrutura crustal da Zona Transversal da Província Borborema.
74e85. http://dx.doi.org/10.1016/j.jsames.2012.12.009. Universidade de S~ ao Paulo, Nordeste do Brasil.
Assumpç~ ao, M., Ferreira, J., Barros, L., Bezerra, H., França, G.S., Barbosa, J.R., Lloyd, S., van der Lee, S., França, G.S., Assumpç~ ao, M., Feng, M., 2010. Moho map of
Menezes, E., Carlos Ribotta, L., Pirchiner, M., Nascimento, A., do, Dourado, J.C., South America from receiver functions and surface waves. J. Geophys. Res. 115,
Talwani, P., 2014. Intraplate seismicity in Brazil. In: Intraplate Earthquakes. B11315. http://dx.doi.org/10.1029/2009JB006829.
Cambridge University Press, Cambridge, pp. 50e71. http://dx.doi.org/10.1017/ Mantovani, M.S.M., Quintas, M.C.L., 1996. A contribution to the study of the Parana 
CBO9781139628921.004. basin evolution. In: Abstracts, 30th International Geological Congress. Beijing,
Assumpç~ ao, M., Schimmel, M., Escalante, C., Roberto Barbosa, J., Rocha, M., p. 102.
Barros, L.V., 2004. Intraplate seismicity in SE Brazil: stress concentration in Mantovani, M.S.M., Quintas, M.C.L., Shukowsky, W., Brito Neves, B.B., 2005. De-
lithospheric thin spots. Geophys. J. Int. 159, 390e399. http://dx.doi.org/10.1111/ limitation of the Paranapanema Proterozoic block: a geophysical contribution.
j.1365-246X.2004.02357.x. Episodes 28, 18e22.
Bahia, R.B.C., Martins-Neto, M.A., Barbosa, M.S.C., Pedreira, A.J., 2007. Ana lise da Matos, R.M.D. de, Brown, L.D., 1992. Deep seismic profile of the amazonian craton
evoluç~ ao tectonossedimentar da Bacia dos Parecis atrave s de me todos poten- (northern Brazil). Tectonics 11, 621e633. http://dx.doi.org/10.1029/91TC03091.
ciais. Rev. Bras. Geocie ^ncias 37, 639e649. Mooney, W.D., Ritsema, J., Hwang, Y.K., 2012. Crustal seismicity and the earthquake
Balmino, G., Vales, N., Bonvalot, S., Briais, A., 2012. Spherical harmonic modelling to catalog maximum moment magnitude (Mcmax) in stable continental regions
ultra-high degree of Bouguer and isostatic anomalies. J. Geod. 86, 499e520. (SCRs): correlation with the seismic velocity of the lithosphere. Earth Planet.
http://dx.doi.org/10.1007/s00190-011-0533-4. Sci. Lett. 357e358, 78e83. http://dx.doi.org/10.1016/j.epsl.2012.08.032.
Berrocal, J., Marangoni, Y., de S a, N.C., Fuck, R., Soares, J.E.P., Dantas, E., Perosi, F., Musacchio, G., Mooney, W.D., Luetgert, J.H., Christensen, N.I., 1997. Composition of
Fernandes, C., 2004. Deep seismic refraction and gravity crustal model and the crust in the grenville and appalachian provinces of north America inferred
tectonic deformation in Tocantins Province, Central Brazil. Tectonophysics 388, from Vp/Vs ratios. J. Geophys. Res. Solid Earth 102, 15225e15241. http://
187e199. http://dx.doi.org/10.1016/j.tecto.2004.04.033. dx.doi.org/10.1029/96JB03737.
Bianchi, M.B. De, 2008. Variaço ~es da estrutura da crosta, litosfera e manto para a Nascimento, P.F. do, 2011. Invers~ ao de dados geofísicos do Brasil Central. Uni-
plataforma Sul Americana atrave s de funço ~es do receptor para ondas P e S. versidade de Brasília.
Universidade de Sa ~o Paulo, Sa ~o Paulo. http://dx.doi.org/10.11606/T.14.2008.tde- Niu, F., Bravo, T., Pavlis, G., Vernon, F., Rendon, H., Bezada, M., Levander, A., 2007.
22092008-121708. Receiver function study of the crustal structure of the southeastern Caribbean
Christensen, N.I., Mooney, W.D., 1995. Seismic velocity structure and composition of plate boundary and Venezuela. J. Geophys. Res. 112, B11308. http://dx.doi.org/
the continental crust: a global view. J. Geophys. Res. Solid Earth 100, 10.1029/2006JB004802.
9761e9788. http://dx.doi.org/10.1029/95JB00259. Pav~ao, C.G., França, G.S., Marotta, G.S., Menezes, P.H.B.J., Neto, G.B.S., Roig, H.L., 2012.
Clayton, R.W., Wiggins, R.A., 1976. Source shape estimation and deconvolution of Spatial interpolation applied a crustal thickness in Brazil. J. Geogr. Inf. Syst. 4,
teleseismic bodywaves. Geophys. J. Int. 47, 151e177. http://dx.doi.org/10.1111/ 142e152. http://dx.doi.org/10.4236/jgis.2012.42019.
j.1365-246X.1976.tb01267.x. Pimentel, M.M., Fuck, R.A., 1992. Neoproterozoic crustal accretion in central Brazil.
Cordani, U.G., Tassinari, C.C.G., Teixeira, W., Basei, M.A.S., Kawashita, K., 1979. Geology 20, 375. http://dx.doi.org/10.1130/0091-7613(1992)020<0375:
Evoluç~ ^ nica da Amazo
ao tecto ^nia com base nos dados geocronolo gicos. In: Actas. NCAICB>2.3.CO;2.
Instituto de Investigaciones Geologicas, Ciudad de Arica, pp. J137eJ148. Reguzzoni, M., Sampietro, D., 2015. GEMMA: an Earth crustal model based on GOCE
Costa, M.M.D., França, G.S., 2008. Estudo da estrutura da crosta na regia ~o norte do satellite data. Int. J. Appl. Earth Obs. Geoinf 35, 31e43. http://dx.doi.org/10.1016/
brasil. In: Estudo Da Estrutura Da Crosta Na Regia ~o Norte Do Brasil. Brasília, j.jag.2014.04.002.
pp. 1e8. Rosa, J.W.C., Rosa, J.W.C., Fuck, R.A., 2016. The structure of the Amazonian craton:
Daly, M.C., Andrade, V., Barousse, C.A., Costa, R., McDowell, K., Piggott, N., Poole, A.J., available geophysical evidence. J. South Am. Earth Sci. 70, 162e173. http://
2014. Brasiliano crustal structure and the tectonic setting of the Parnaíba basin dx.doi.org/10.1016/j.jsames.2016.05.006.
of NE Brazil: results of a deep seismic reflection profile. Tectonics 33, Rosa, J.W.C., Rosa, J.W.C., Fuck, R.A., 2012. Crust and upper mantle structure in
2102e2120. http://dx.doi.org/10.1002/2014TC003632. central Brazil derived by receiver functions and SKS splitting analysis. J. South
Faria, H.P.A., 2015. Caracterizaça ~o de domínios Tectono-Geofísicos na Bacia dos Am. Earth Sci. 34, 33e46. http://dx.doi.org/10.1016/j.jsames.2011.09.001.
Parecis com base em dados de me todos potenciais. Universidade de Brasília. Santos, J.O.S., 2003. Geotecto ^nica dos escudos das Guianas e Brasil-Central. In:
França, G.S.L.A. de, 2003. Estrutura da crosta no Sudeste e Centro-Oeste do Brasil, Bizzi, L.A., Schobbenhaus, C., Vidotti, R.M., Gonçalves, J.H. (Eds.), Geologia,
usando funça ~o do receptor. Universidade de Sa ~o Paulo. Tecto^nica E Recursos Minerais Do Brasil (CPRM). Universidade de Brasília,
Fuck, R.A., Brito Neves, B.B., Schobbenhaus, C., 2008. Rodinia descendants in South Brasília, pp. 169e226.
America. Precambrian Res. 160, 108e126. http://dx.doi.org/10.1016/ Smith, W.H.F., Wessel, P., 1990. Gridding with continuous curvature splines in
j.precamres.2007.04.018. tension. GEOPHYSICS 55, 293e305. http://dx.doi.org/10.1190/1.1442837.
Hasui, Y., Haralyi, N.L.E., Schobbenhaus, C., 1984. Elementos geofísicos e geolo gicos Soares, J.E., Berrocal, J., Fuck, R.A., Mooney, W.D., Ventura, D.B.R., 2006. Seismic
da Regia ~o Amazo ^nica: subsídios para o modelo geodina ^mico. In: II Symposium characteristics of central Brazil crust and upper mantle: a deep seismic
Amazo ^nico. Departamento Nacional da Produça ~o Mineral, Manaus, refraction study. J. Geophys. Res. Solid Earth 111. http://dx.doi.org/10.1029/
pp. 129e147. 2005JB003769.
James, D.E., Assumpça ~o, S. de, Snoke, J.A., Ribotta, L.C., Kuehnel, R., 1993. Preliminary Trabant, C., Hutko, a. R., Bahavar, M., Karstens, R., Ahern, T., Aster, R., 2012. Data
seismic studies of continental lithosphere beneath se Brazil. Eos Trans. Am. products at the IRIS DMC: stepping stones for research and other applications.
Geophys. Union 426. Seismol. Res. Lett. 83, 846e854. http://dx.doi.org/10.1785/0220120032.
Juli
a, J., Assumpç~ ao, M., Rocha, M.P., 2008. Deep crustal structure of the Parana  Basin Trindade, C.R., Soares, J.E.P., Fuck, R.A., Chatack Carmelo, A., Peixoto, C.L.O., 2014.
from receiver functions and Rayleigh-wave dispersion: evidence for a frag- Estrutura crustal do brasil central. Comun. Geol. 101, 339e343.
mented cratonic root. J. Geophys. Res. 113, B08318. http://dx.doi.org/10.1029/ Wessel, P., Smith, W.H.F., Scharroo, R., Luis, J., Wobbe, F., 2013. Generic mapping
2007JB005374. Tools: improved version released. Eos. Trans. Am. Geophys. Union 94, 409e410.
Krüger, F., Scherbaum, F., Rosa, J.W.C., Kind, R., Zetsche, F., Ho € hne, J., 2002. Crustal http://dx.doi.org/10.1002/2013EO450001.
and upper mantle structure in the Amazon region (Brazil) determined with Zhu, L., Kanamori, H., 2000. Moho depth variation in southern California from
broadband mobile stations. J. Geophys. Res. Solid Earth 107. http://dx.doi.org/ teleseismic receiver functions. J. Geophys. Res. Solid Earth 105, 2969e2980.
10.1029/2001JB000598. ESE 17-1-ESE 17-12. http://dx.doi.org/10.1029/1999JB900322.