Accepted Manuscript

Quantitative seismic geomorphology of a submarine channel system in SE Brazil

(Espírito Santo Basin): Scale comparison with other submarine channel systems

Yongpeng Qin, Tiago M. Alves, José Constantine, Davide Gamboa

PII: S0264-8172(16)30329-4
DOI: 10.1016/j.marpetgeo.2016.09.024
Reference: JMPG 2684

To appear in: Marine and Petroleum Geology

Received Date: 24 December 2015

Revised Date: 5 September 2016
Accepted Date: 22 September 2016

Please cite this article as: Qin, Y., Alves, T.M., Constantine, J., Gamboa, D., Quantitative seismic
geomorphology of a submarine channel system in SE Brazil (Espírito Santo Basin): Scale comparison
with other submarine channel systems, Marine and Petroleum Geology (2016), doi: 10.1016/
j.marpetgeo.2016.09.024.

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
Quantitative seismic geomorphology of a submarine channel system in SE Brazil

(Espírito Santo Basin): scale comparison with other submarine channel systems

Yongpeng Qin1, Tiago. M. Alves1, José Constantine2, Davide Gamboa3

1) 3D Seismic Lab, School of Earth and Ocean Sciences, Cardiff University – Main Building, Park

PT
Place, Cardiff, CF10 3AT, United Kingdom
2) Geosciences Department, Williams College – Clark Hall, 947 Main Street, Williamstown,
MA01267, United States of America

RI
3) British Geological Survey – Cardiff Office, Columbus House, Village Way, Greenmeadow
Springs, Tongwynlais, Cardiff CF15 7NE, United Kingdom

SC
Abstract
Detailed morphological analysis of a Pleistocene-Quaternary submarine channel system in terms of hierarchical
framework, including channel and valley scale, were carried out using a 3D seismic volume from offshore

U
Espírito Santo, SE Brazil. The channel morphology shows marked variations along the full channel length. For
AN
example, the cross-sectional area of the channel reduces by a factor of 70 in 10 km. The changes of channel
morphology may have related to temporal and spatial variations in flow volume within the channel. In the same
channel system, the valley shows three distinct segments (Segments A to C), with similar aspect ratio but marked
M

variation in morphology along the valley distance. Segment B is characterized by the largest valley-base width,
valley width, and cross-sectional area compared to other segments. Such valley enlargement in Segment B was
resulted from relatively higher degree of lateral channel migration, which lead to the widening of both valley base
D

and valley. Spatial variations in erosion processes are observed along the valley. Segment B is mainly widened
through cut bank erosion during lateral channel migration, while Segment C is characteristic by inner bank
TE

erosion in the form of various sized failure scars. Spatial variations in both channel and valley morphology in the
same channel system suggests an important role of local factors (e.g. salt diapirs, tributaries, overbank collapses)
on the development of the channel system. Detailed morphological analysis of the channel system provide an
EP

analogue for studying channels and valleys in other continental slopes.

Key words: submarine channel, hierarchical framework, geomorphology, SE Brazil.

C

Highlights:
AC

The channel and valley show marked variation of morphology in the same channel system.

The variations in channel morphology may have related to spatial and temporal variations in flow discharge.

The variations in valley morphology is influenced by the erosional processes within the valley.

1
 ACCEPTED MANUSCRIPT
1. Introduction bridge the gap between seismic and outcrop data, more
morphological data are needed from seismic data, especially
Submarine channels are major conduits for sediment the spatial variation of channel size in different scales, which
transported from continental shelves to abyssal plains, and their are less documented in previous studies.
deposits comprise important reservoirs in deep-water This work focuses on a submarine channel system on the
sedimentary basins when filled with sand- or silt-prone modern seafloor, the Rio Doce channel system developed on
sequences (Mayall and Stewart, 2000; Mayall et al., 2006; the continental slope of SE Brazil (Fig. 1). This study
Wynn et al., 2007). To better organize and compare documents the detailed morphological characteristics of the
architectural elements associated with submarine channels, channel system in both channel and valley scale, and is

PT
hierarchical frameworks have been proposed by several authors followed by a scale comparison with other channel systems in
(Gardner and Borer, 2000; Campion et al., 2000; Sprague et al., terms of hierarchical framework.
2002, 2005). Various scale stratigraphic elements, such as

RI
channel element, channel complex and channel complex set 2. Terminology
have been extensively recorded from previous studies (e.g.

SC
Gardner et al., 2000, 2003; Abreu et al., 2003; Deptuck et al., Three main types of morphological features (i.e. channel,
2003; McHargue et al., 2011; Di Celma et al., 2011; Thomas channel belt and valley) and their deposits are defined in the
and Bodin, 2013; Bain and Hubbard, 2016) following section (Fig. 2).
Availability of high-resolution seismic and bathymetric The term submarine channel, or channel as used herein, is

U
data has led to the detailed morphological studies of submarine used to define V or U shaped negative features on the seafloor
AN
channel systems in recent years, such as Amazon fan (Flood et (Fig. 2c). It is a conduit transferring sediments from continental
al., 1987; Pirmez and Flood, 1995), offshore Gabon (Wonham shelf to the deep sea. A submarine channel corresponds to the
et al., 2000), Zaire fan (Babonneau et al., 2002), Niger delta channel element in terms of established stratigraphic
M

slope (Deptuck et al., 2007; Jobe et al., 2015) and Caribbean hierarchical frameworks (Sprague et al., 2002; Abreu et al.,
area (Wood and Mize-spansky, 2009). However, the majority of 2003; McHargue et al., 2011; Thomas and Bodin, 2013).
previous studies were focused on either channel scale (e.g. Buried channels refer to subsurface channels filled by sediment
D

Flood et al., 1987; Clark et al., 1992; Pirmez and Flood, 1995; that differentiate from the channel on the seafloor. Depositional
Babonneau et al., 2002; Ferry et al., 2005; Antobreh et al., 2006; units within subsurface channel comprise channel fill. The
TE

Gee et al., 2007; Estrada et al., 2005; Clark and Cartwright, lowest point of channels is defined as the channel thalweg.
2009; Peakall et al., 2012; Kolla et al., 2012; Georgiopoulou Channel belt is defined as a morphological feature on the
and Cartwright, 2013; Jobe et al., 2015) or channel complex seafloor. It is composed of channel and terraces (Fig. 2d). A
EP

scale (e.g. Caterall et al., 2010; Wood and Mize-spansky, 2009), terrace is defined as a topographic flat bordering a channel
with few of them focusing on the morphological characteristics within the channel belt (Babonneau et al., 2004). Terraces can
of both scales in a same channel system (Wonham et al., 2000; be formed various mechanisms, such as entrenchment, bank
C

Deptuck et al., 2007) failure, deposition of inner/internal levees, lateral channel

Meanwhile, outcrop studies of submarine channels have migration, or meander cut-offs (Babonneau et al., 2004;
AC

documented various stratigraphic elements, from small-scaled Deptuck et al., 2003; Hansen et al., 2015). The term channel
channel element to large scaled channel complex and channel belt is used by many studies (Deptuck et al., 2003; Posamentier
complex set (e.g. Grecula et al., 2003; Lien et al., 2003; et al., 2003; Catterall et al., 2010; Kane et al., 2011; Gamberi et
Vigorito et al., 2006; Hubbard et al., 2008, 2009, 2014; Thomas al., 2013; Hansen et al., 2015), the similar architectural feature
and Bodin, 2013; Gamberi et al., 2013; Macauley and Hubbard, in these studies refers to valley in this work.
2013; Bain and Hubbard, 2016). The size of those channel Valley is defined as a higher order architectural element
complex and channel complex set are similar to seismic scale than the channel (Fig. 2e) (Sameul et al., 2003; Kolla et al.,
and allow comparisons between seismic and outcrop data 2007; Janocko et al., 2013). It comprises subsurface valley fill
become possible (e.g. Lien et al., 2003; Vigorito et al., 2006; and channel belt on the seafloor (Fig. 2e). The valley
Thomas and Bodin, 2013; Bain and Hubbard, 2016). To further corresponds to a channel complex when considering

2
 ACCEPTED MANUSCRIPT
stratigraphic hierarchical framework, both of which are syn-rift stage developed from the Late Berriasian to Early
composed of two or more channel-fill episodes genetically Aptian and was dominated by intense tectonic subsidence. This
related to each other (Sprague et al., 2002; Abreu et al., 2003; stage records the widespread formation of rift basins in the
McHargue et al., 2011; Thomas and Bodin, 2013). The valley is ESB, in which deposition of fluvial and lacustrine sediments
formed by lateral migration and vertical stacking of a single took place (Fig. 3). Magmatic activity also occurred in this
channel element (e.g. Sameul et al., 2003; Deptuck et al., 2003; stage (Fig. 3), resulting in basaltic and andesitic volcanism
Babonneau et al.,2004; Kolla, 2007; Jaocko et al., 2013), and (Ojeda, 1982). An angular unconformity marks the boundary
has a composite erosional surface consisting of a valley base between the syn-rift and post-rift stages (Chang et al., 1992).
and valley walls (Fig. 2e) (erosional fairway sensu Deptuck et Following syn-rift tectonics, the transitional stage spanned

PT
al., 2003 and Catterall et al., 2010; master erosion/incision Middle Aptian to Late Aptian/Early Albian, in which the
surface sensu Gee et al., 2007; channel-belt erosion surface deposition of thick evaporite sequences and marine carbonates
sensu Kane and Hodgson, 2011; bounding erosion surface predominated (Ojeda, 1982).

RI
sensu Hansen et al., 2015). The lowest point of the valley’s The drift phase lasted from Late Aptian/Early Albian to
erosional surface is the valley thalweg. present-day, and open-marine strata predominated throughout

SC
Levees are overbank deposits with gull-wing shaped this stage (Fig. 3) (Ojeda, 1982; Chang et al., 1992). Strata
geometries that converge away from the channel thalweg. deposited at this time is divided into two megasequences:
External levees (outer levees sensu Deptuck et al., 2003, 2007; early-drift transgressive and late-drift regressive (Fig. 3)
master-bounding levees sensu Posamentier, 2003) and internal (Mohriak, 2003; Fiduk et al., 2004). The transgressive

U
levees (inner levee sensu Deptuck et al., 2003, 2007) are used megasequence consists of Albian carbonate platforms below
AN
to differentiate levees deposited outside and inside the valley muddy and sandy turbidites, which mark a general
(Kane et al., 2011). deepening-upwards trend towards the end of the Cretaceous
The term channel system is used here to refer to all (Davison, 1999; Fiduk et al., 2004; Alves et al., 2009). The
M

genetically related morphological and depositional components. regressive megasequence is associated with the reactivation of
It is defined by the valleys (or channel complex) plus the rift structures and episodic magmatic activity (Demercian et al.,
external levess confining them. Two or more genetically-related 1993; Cobbold et al., 2001; Mohriak et al., 2008), and records
D

channel systems comprises channel complex set. the emplacement of the Abrolhos bank to the north of the study
area (Cordani, 1970). A large amount of sediment sourced from
TE

3. Geological setting the Abrolhos bank and onshore rivers was transported
downslope by submarine canyons and channels during the
3.1 Tectono-sedimentary evolution of the regressive megasequence (Bruhn and Walker, 1997; Davison et
EP

Espírito Santo Basin al., 1999).

The Espírito Santo Basin (ESB) is located on the SE 3.2 Salt tectonics in the Espírito Santo Basin
C

Brazil continental margin between the Abrolhos bank and

Campos Basin (Fig. 1a). The width of ESB’s continental shelf Salt tectonic deformation has been active in the ESB since
AC

increases from 50-60 km in the south to 240 km on the the Albian (Fiduk et al., 2004). The main triggers for salt
Abrolhos Bank in the north (Bastos et al., 2015). The modern tectonics in the region are gravitational gliding and differential
shelf break varies in depth from 40 to 80m in the basin sediment loading by overburden rocks of different thickness
(Knoppers et al., 1999; Bastos et al., 2015). (Demercian et al., 1993; Fiduk et al., 2004). As a result, three
Four tectono-sedimentary stages have been interpreted in tectonic domains are interpreted in the ESB, each showing
the ESB: pre-rift, syn-rift, transition and drift stages (Fig. 3) different salt structural styles (Fig. 4) (Fiduk et al., 2004). The
(Ojeda, 1982). The pre-rift stage spans the Late Jurassic to most proximal domain is characterized by salt rollers and rafts,
Early Cretaceous, and records the deposition of and salt deformation is dominated by extensional processes
fluvial-lacustrine sediments in basins dominated by moderate (Demercian et al., 1993; Fiduk et al., 2004; Alves, 2012). In the
fault-related subsidence (Ojeda, 1982; Chang et al., 1992). The transitional mid-slope domain, where the study area is located,
salt diapirs predominate (Fiduk et al., 2004). Basinwards, the
3
 ACCEPTED MANUSCRIPT
compressional domain is dominated by allochthonous salt
canopies, tongues and overhangs (Fiduk et al., 2004; Mohriak 4.1 Seismic dataset
et al., 2008).
In the study area, several salt diapirs are rooted on two The interpreted 3D seismic volume is located at the
NW-SE trending salt ridges (Gamboa and Alves, 2015). Some northern Espírito Santo basin (Fig. 1a). The data were acquired
of these salt diapirs, such as D1, D3, D4, D5 and D6, have a by a dual airgun array and a 6 x 5700 m array of streamers. The
seafloor expression at present (Fig. 1b). Various architectural survey has a bin spacing of 12.5 by 12.5 m and a 2 ms vertical
elements, such as mass-transport deposits (MTDs), turbidite sampling interval. Data processing included resampling,
lobes, submarine canyons and channels are observed within the spherical divergence corrections, and zero-phase conversions

PT
salt-withdrawal basin delimited by salt diapirs D1 to D6 undertaken prior to stacking, 3-D prestack time migration using
(Gamboa and Alves, 2015). These elements are strongly the Stolt algorithm, and one-pass 3-D migration. The vertical
influenced by the relative location and movement of the salt resolution of the data is ~10 m at the depth of analysis in this

RI
diapirs (Baudon and Cartwright, 2008; Alves et al., 2009; study, based on dominant frequency of 40 Hz and a P-wave
Gamboa et al., 2012; Gamboa and Alves, 2015). Topographic velocity of 1600 m/s for near-seafloor strata. A water-column

SC
confinement created by salt diapirs is reflected as changes in velocity of 1480 m/s was used to depth-convert the seafloor.
channel density, geometry and sinuosity (Alves et al., 2009; The polarity of data is SEG normal, i.e., positive amplitude
Gamboa et al., 2012). As an example, channel density reflections (red) on the seismic profiles represent an increase in
decreases from the unconfined pre-confluence region to the acoustic impedance (Fig. 2a).

U
topographically confined confluence region of the Rio Doce
AN
channel system (Gamboa et al., 2012). 4.2 Morphological analysis of the channel
system
3.3 Rio Doce Canyon System in the Espírito
Santo Basin
M

The mapping of the seafloor and channels was

accomplished using line-by-line interpretation, involving a
The Rio Doce Canyon system, defined as a set of Eocene combination of manual- and auto-tracking. A quantitative
D

to recent stacked submarine channels that developed in the analysis of the channel system was made following the
northern part of the ESB, is one of the most important methods of Deptuck et al. (2007), including channel and valley
TE

depositional features offshore SE Brazil (Fiduk et al., 2004). measurements (Fig. 6).
The sediment source of the canyon system is affected by the Channel measurements include the depth of channel
longshore drifts along the SE Brazil, which play an important thalweg and channel bank below sea level, channel height,
EP

role in distributing sediment on the continental shelf channel width, the aspect ratio (width/height) and
(Dominguez et al., 1992; Martin and Suguio, 1992). At least cross-sectional area (CSA) (Fig. 6c). These parameters were
three major channel events have been identified within the Rio measured in equally spaced (125 m) cross-sections
C

Doce Canyon system (Fig. 5): (1) an Eocene channel system, perpendicular to the channel axis line. Channel gradient was
which was controlled by adjacent faults generated on the flanks calculated based on depth changes along the channel distance.
AC

of growing diapirs (Alves et al., 2009), (2) a Miocene channel Valley measurements were conducted at cross-sections
system with multiple laterally shifting channels, and (3) a perpendicular to the valley axis at 65 m intervals. The
Pliocene-Quaternary channel complex set majorly confined by parameters measured include the depth of valley thalweg and
salt diapirs (Gamboa et al., 2012). The studied late valley bank below sea level, valley height, valley width, the
Pleistocene-Quaternary channel system belongs to the aspect ratio (width/height) and cross-sectional area (CSA) of
Pliocene-Quaternary channel complex set, which is bounded at the valley (Fig. 6d). The gradient of valley thalweg and valley
its base by an intra-Pliocene unconformity of regional bank were calculated based on depth changes along the valley
expression (Fig. 5) (França et al., 2007). distance.

4. Data and Methodology 5. Seafloor morphology

4
 ACCEPTED MANUSCRIPT
patterns between the east tributary and the post-confluence
The study area is located on the southern flank of the channel, as well as the continuity of the channel thalweg,
Abrolhos bank, which is associated with a unique slope indicates that these two channel segments constitute the main
configuration in the ESB (Fig. 1a). The presence of the flow pathway at present (Gamboa et al., 2012). Other small
Abrolhos bank forces a shift of the orientation of continental tributaries connect the east tributary are also observed on the
shelf, changing from NE-SW striking in the southwestern parts seafloor (Fig. 9). Both west and east tributaries change their
to ENE-WSW striking in the northeastern part of the Espírito orientation in the pre-confluence region, and were diverted into
Santo Basin, resulting in two different slope trends in the study the confluence region due to salt diapirs (Fig. 9). The general
area (Fig. 1b). The western half of study area displays a orientation of the channel changes to nearly N-S until the

PT
decreasing slope from 8° in northwest to 1° in southeast, southern boundary of the seismic volume is reached (Fig. 9).
whereas the eastern half shows a gentler slope ranges from 1°-2° The dominant flow pathway (east tributary and
(Figs. 1b and 7). The seafloor slope configuration is also post-confluence channel) is the focus in the following sections

RI
affected locally by growing salt diapirs, and in some areas, the (Fig. 9). Along that pathway, the channel system is divided into
down-to-basin slope angle is reduced to nearly 0°, and different segments (Fig. 10), based on the morphological
increased basinward of the diapirs (Fig. 7).

SC
analysis conducted at channel (Fig. 11 and Table 1) and valley
Three slope regions have been defined based on the degree scales (Fig. 12 and Tables 2).
of confinement imposed by salt diapirs deforming the seafloor
(Fig. 8). These directly relate to the channel confluence regions 6.1 Quantitative channel analysis

U
in the study area (Gamboa et al., 2012). A pre-confluence slope
AN
region (Zone 1) is relatively unconfined and shows a variety of The seafloor channel is 42 km long within the seismic
erosional features, such as gullies, channels, irregular dataset. It starts from a water depth of ~1000 m, down to ~1700
depressions and headwalls of mass-wasting events (Fig. 8). The m at the southern edge of the seismic volume. The channel
confluence region (Zone 2) presents a relatively higher
M

shows marked variations in its morphology (Fig. 11 and Table

topographic confinement when compared to Zone 1 due to the 1). Five distinct segments (Segments a to e) are recognized
presence of salt diapirs D2, D3, D5 and D6. As a result, based on changes in their morphology (Figs. 10, 11 and Table
D

turbidity flows sourced from upslope areas are diverted by 1).

these salt diapirs, as evidenced by the shift in orientation of two
TE

tributaries and the confluence point on the seafloor channel 6.1.1 Depth profile of channel and channel height
(Fig. 8). For example, the pathway of east tributary changes The depth profile of the channel thalweg shows an
from NNE-SSW in Zone 1 to NE-SW in Zone 2 due to the exponential trend and is divided into three intervals based on
EP

presence of salt diaper D5 (Fig. 8). A distinct confluence point the variations in channel gradient (Fig. 11a). Channel gradient
for two Miocene tributaries has been observed in the same is of 1.47° (25.7 m/km) in the first interval, where the steepest
region beneath the seafloor (Gamboa et al., 2012), suggesting parts of the channel are recorded (Fig. 11a). Between 12 km
C

that the present-day slope configuration was established in and 27 km in interval 2, the channel depth shows a stepped
Miocene. Downslope from the post-confluence region (Zone 3), profile and the channel gradient decreases to 0.83° (14.6 m/km)
AC

the seafloor becomes less confined after diapirs D2 and D5, and due to the presence of salt diapir D5. In the remainder of the
it is relatively smooth with fewer erosional features (Fig. 8). channel (interval 3), the channel gradient decreases to 0.71°
(12.3 m/km) between 27 and 37 m, and reaches the lowest
6. General morphology of the late value of 0.5° (8.7 m/km) in the last 8 km (Fig. 11a).
Pleistocene-Holocene channel system The depth profile of channel bank shows marked
variations along the channel (Fig. 11a), which correlate with
The late Pleistocene-Quaternary channel system is only variations in the channel height profile (Fig. 11b). There are
partially filled, and includes a sinuous channel on the seafloor three intervals observed in the depth of channel bank and
(Figs. 5 and 6). The seafloor channel comprises two upslope channel-height profiles (Figs. 11a and 11b). The first interval (0
tributaries in Zone 1 and a post confluence channel downslope – 12 km) has the highest value of channel height, which ranges
in Zones 2 and 3 (Fig. 9). The continuity of sedimentary fill
5
 ACCEPTED MANUSCRIPT
from 80 m and 156 m (Fig. 11b). In interval 2, between 12 and km, where the CSA decreases to 0.025 km2 (Fig. 11e). Between
27 km, the channel height decreases rapidly due to the reduced 10 km and 25 km, the CSA reduces by a factor of 33, from
depth of channel bank (Figs. 11a and 11b). The channel height 0.067 km2 at 10 km to its minimum value of 0.002 km2 at 21
reduces to its minimum value of 13 m at 21 km, and then km. This significant reduce is followed by a threefold increase
fluctuates between 22 m and 58 m for the remainder of the from 21 to 22 km, where the CSA is 0.008 km2 and remains
interval (Fig. 11b). The third interval starts with a marked constant for the next 3 km. The CSA shows a rapid rise from 25
increase in channel height from 22 m to 100 m at 27 km due to km and an increase to 0.035 km2 at 27 km. It then reduces
an abrupt decrease in the depth of channel bank, and is gradually to 0.013 km2 at 34 km, and varies between 0.009 km2
followed by a decreasing trend. The channel height reduces to and 0.02 km2 for the remainder of the channel (Fig. 11e)

PT
26 m at the southern edge of the seismic volume (Fig. 11b).
6.1.4 Variations in channel morphology
6.1.2 Width profile of channel and channel floor The channel displays considerable variation of

RI
The channel-width profile displays four intervals (Fig. morphology along the channel distance (Fig. 11 and Table 1).
11c). It fluctuates between 445 and 1061 m in the first 13 km Five segments (Segment a to e) are observed according to

SC
within interval 1. In interval 2, the channel width decreases variations in channel morphological parameters (Figs. 10 and
from 689 m to its lowest value of 243 m at 27 km. This 11, Table 1).
decreasing trend is followed by an increase of channel width In Segment a for the first 13 km, the channel is steepest
between 27 km and 32 km (interval 3) where the channel width and has the largest cross sections, as shown by the highest

U
ranges from 403 m to 610 m. In interval 4, the channel width value of channel gradient, height, width and CSA (Fig. 11 and
AN
reduces rapidly to 310 m at 32 km and then rises progressively. Table 1). In Segment b between 13 km and 22 km, the channel
It increases to 440 m at the southern limit of the seismic gradient reduces and is accompanied by significant decreases
volume (Fig. 11c) of channel height, width and CSA (Fig. 11 and Table 1). For
M

The width of channel floor varies from 80 to 400 m, with example, the channel width reduces more than 200 m and the
an average value of 203 m, and its profile shows four intervals CSA decreases by a factor of 23 in this segment. However, the
(Fig. 11c). For the most of channel length, the channel-floor width of the channel floor and the aspect ratio increase to their
D

width shows slight variations, but it displays two increasing maximum value against reduced channel size in Segment b
trends starting at 12 km in interval 2 and 36 km in interval 3 (Figs. 11b – 11e and Table 1).
TE

(Fig. 11c). These increases correspond to the decrease of In Segment c (22 – 27 km), the channel size decreases to
channel gradient at 12 km and 36 km (Figs. 11a and 11c). its lowest value, with lowest channel height of 38 m,
channel-floor width of 165 m, channel width of 345 m and
EP

6.1.3 Aspect ratio (width/height) and cross-sectional area (CSA) CSA of 0.008 km2 (Table 1). The aspect ratio of the channel
of channel declines to 9 in this segment (Table 1).
The aspect ratio of the channel ranges from 5 to 27 and In Segment d between 27 km and 32 km, abrupt increases
C

four intervals are observed along the channel (Fig. 11d). It in channel height and width are observed, and are accompanied
shows small changes in the first 13 km, with an average value by a fourfold increase of the channel CSA (Figs. 11b, 11c, 11e
AC

of 6 in interval 1. The ratio rises in interval 2, between 13 km and Table 1). In Segment e (32 – 42 km), the channel width,
and 21 km, and increases to its maximum value of 27 at 20.75 height and CSA decreases again with increased aspect ratio,
km. The third interval starts with an abrupt reduction at 21 km, which resulted from increased width and decreased channel
where the aspect ratio reduces to 9 and then remains nearly height in this segment (Figs. 11b - 11e and Table 1).
constant. Between 27.5 and 36.5 km, the ratio within interval 4
displays an increasing trend towards the southern edge of the 6.2 Quantitative valley analysis
seismic volume, where the ratio increases up to 17 (Fig. 11d).
The CSA of the channel shows the similar trends with The valley is divided into three segments (Segment A to C)
channel height and width (Figs. 11b, 11c and 11e). The CSA based on changes in valley morphology (Figs. 10, 12 and Table
2
is ～0.06 km in the first 10 km except an abrupt reduction at 2 2). It shows various architectural elements (e.g. terraces and

6
 ACCEPTED MANUSCRIPT
scars) in different segments (Fig. 10). interval (24 – 35 km), the valley width varies from 1000 m to
1500 m for the most part of this interval, with two peak values
6.2.1 Depth profile of valley thalweg and valley wall, and of 2187 at 26.4 km and 2755 m at 33.25 km. The latter
valley height represents where the valley width reaches its maximum value
There are five intervals observed in the depth profile of (Fig. 12c).
valley thalweg (Fig. 12a). The valley is steepest in the first 6 The width of valley base varies from 161 m to 700 m for
km, with a gradient of 1.88° (32.9 m/km). The valley gradient the most part of the valley. However, a rapid increase occurs
decreases to 1.34° (23.5 m/km) between 6 km and 13 km due to between 19 km and 23 km, where the valley-base can be up to
the presence of salt diapir D5. For the next 3 km, the valley 1738 m wide and coincides with an increase in valley width

PT
thalweg is difficult to recognize because of its similar seismic (Fig. 12c).
facies with MTD beneath, both of which are composed of
discontinuous, high amplitude reflections. Between 16 km and 6.2.3 Aspect ratio and cross-sectional area (CSA) of valley

RI
22 km, the valley gradient declines to 1.04° (18.2 m/km), and The aspect ratio of the valley has a similar trend to the
then is followed by an increase to 1.39° (24.2 m/km). In the valley width (Figs. 12c and 12d), suggesting the valley width

SC
remainder of the valley, the valley gradient reduces to the varies much more than the valley height. The aspect ratio
lowest gradient of 0.65° (11.4 m/km) between 30 m and 35 m ranges from 3 to 5 in the first 20 km. It is followed by an
(Fig. 12a). increase in interval 2 (20 – 24 km), where the aspect ratio rises
The depth profile of the valley wall has the steepest up to 9. The aspect ratio decreases to 4 for the majority of the

U
interval in the first 4 km, with a gradient of 2.01° (35 m/km). third interval (24 – 35 m) but with two peak values, 7 at 26.5 m
AN
This interval is followed by a decreased valley-wall gradient of and 9 at 33.5 m, induced by the increased valley width.
1.28° (22.4 m/km) between 4 and 21 km, and 0.72° (12.6 m/km) The valley CSA also share the similar pattern with the
from 21 to 27 km. In the fourth interval between 27 and 35 km, valley width (Figs. 12c and 12e). The CSA is highest between
17 m and 23 m, where it ranges from 0.192 km2 to 0.366 km2,
M

the valley-wall gradient increases to 1.08° (18.8 m/km).

The valley height is correlated to variations in both the while in the first and third intervals, it varies between 0.03 to
depth of valley wall and valley thalweg (Figs. 12a and 12b). It 0.21 km2 (Fig. 12e).
D

ranges from 230 m to 350 m along the full length of the

channel (Fig. 12b). In interval 1, the valley height reduces due 6.2.4 Variation in valley morphology
TE

to a rapid decrease of valley-wall depth (Figs. 12a and 12b). The valley is divided into three segments based on
This trend is followed by an increase of valley height in the variation in its morphology (Figs. 10, 12 and Table 2). These
second interval (4 -21 km), where the valley height changes segments have similar aspect ratios but different valley sizes
EP

from 230 m and 310 m. The valley height reduces to its lowest (Table 2).
value of 230 m at the end of interval 2. In interval 3, the valley The valley is smallest in Segment A, with the lowest
height shows a marked increase. It increases to the highest average value of the valley base width of 382 m, valley width
C

value of 350 m at 29 km and is followed by a decrease in the of 1117 m, valley height of 274 m and CSA of 0.106 km2
remainder of the channel (Fig. 12b). (Table 2). The terraces on the eastern wall of the valley in this
AC

segment are interpreted to result from the deposition of internal

6.2.2 Width profile of valley and valley base levees above slump deposits (Figs. 13a and 13b). Transparent
The width profiles of valley and valley base share similar to low amplitude, chaotic reflections were observed at the
patterns. They both have the highest peak value in interval 2 valley base in this segment (Figs. 13a and 13b). Such seismic
and the relatively lower average values in intervals 1 and 3 (Fig. facies has been previously documented and are interpreted as
12c). slump and debris-flow deposit resultant from mass failures on
The valley width ranges from 732 m to 2755 m (Fig. 12c). the valley wall (Mayall et al., 2006; Gee et al., 2007; Janocko
It fluctuates between 732 m and 1439 m in the first interval (0 - et al., 2013). These slump deposits are observed in the first 8
17 km). This is followed by valley width increases up to 2260 km of Segment A within the valley. A continuity of depositional
m between 17 m and 24 km in interval 2 (Fig. 11c). In the third pattern for such a long distance suggests they may have been

7
 ACCEPTED MANUSCRIPT
transported from upslope, unlikely to be formed a localized were formed by shallow-seated bank failures. The presence of
mass failure on valley wall. The terraces in this segment was some scars spatially coincide with the peak value of the valley
formed by the deposition of internal levees overlying slump width at 26-27 km and 33-34 km (Figs. 10 and 12b), suggesting
deposits (Figs. 13a and 13b) that the valley was mainly widened through shallow-seated
In Segment B, between 15 and 22km, the width of bank failures in this segment. The terraces in this segment was
valley-base and valley increases to their maximum average formed by the deposition of internal levees (Figs. 13e and 13f)
values of 839 m and 1602 m, respectively (Table 2). This is and failure block (Fig. 13g).
especially relevant for the width of the valley base, which
increases by twofold from Segment A to Segment B (Table 2). 7. Discussion

PT
Such increases lead to the highest CSA in segment B (Table 2).
The increase in valley size in Segment B is interpreted to result 7.1 Variations in channel and valley
from cut bank erosion during lateral channel migration. The morphology and possible causes

RI
turbidity flows undercut the lower part of the valley and was
followed by the collapse of overhanging blocks (i.e. cantilever The seafloor channel in this study is comparable to

SC
failure), leading to the lateral widening of the valley, especially channel element recorded from stratigraphic records (Fig. 14).
the valley base. They are all considered as basic architectural elements of stratal
Discontinuous, higher amplitude reflections at the valley hierarchy.
base are interpreted as channel basal lags (Figs. 13c and 13d). The valley is the higher order architectural element in

U
Multiple channel forms are identified at the valley base and terms of hierarchical channel frameworks that is descried in
AN
show the trajectory of lateral channel migrations (Figs. 13c and this work (Fig. 14). It is an integrated result of lateral migration
13d), consistent with observations from Deptuck et al., (2003) and vertical stacking of channel elements through time, as
and a model proposed by Sylvester et al. (2011). Middle to high shown by the shift of channel forms on the valley base (Fig. 13).
M

amplitude, parallel reflections are observed above basal lags This observation is consistent with previous studies, which
(Figs. 13c and 13d), similar reflections are also recorded by have shown that valley base is a diachronous surface or a
Deptuck et al. (2003). It is difficult to identify them as composite erosional surface shaped by multiple erosional
D

abandoned channel fill or overbank deposition (Deptuck et al., events (e.g. Deptuck et al., 2003; Syslvester et al., 2011; Kolla
2003; 2007). Some wedge-shape reflections are interpreted as et al., 2012; Thomas and Bodin, 2013; Macauley and Hubbard,
TE

internal levees (Fig. 13c and 13d). 2013; Bain and Hubbard, 2016; Di Celma et al., 2011)
The architecture of the valley fill interpreted in this work
is similar to the interpretation of Benin-major submarine 7.1.2 Channel morphology
EP

channel system of Niger delta slope (Deptuck et al., 2003). The The morphology of channel cross section shows
presence of channel forms at the valley base suggest abrupt, significant variations along the channel (Fig. 10 and Table 1).
discrete or punctuated lateral channel migration (Deptuck et al., This is illustrated, for example, by the rapid decrease in
C

2007). Similar migration process has also been recorded in channel size from Segment a to Segments b and c, especially
Lucia Chica channel system (Maier et al., 2012). This type of the CSA, which reduces by a factor of 70 from Segment a to
AC

migration is in contrast with the continuous migration process Segment c (Table 1).
revealed by Lateral Accretion Package (LAP) in seismic data Channel gradient is considered as a main control on
(Abreu et al., 2003). turbidity-flow behavior (e.g. Friedmann et al., 2000;
The size of the valley decreases in Segment C, as shown Babonneau et al., 2002; McHargue et al., 2011; Wynn et al.
by the lower values of valley width and CSA, despite the 2012; Stevenson et al. 2013). Spatial correlation between
higher height also observed in this segment (Table 2). Arcuate reduced channel gradient and widened channel floor has been
shaped scars of various sizes of are observed on the bank of the observed at 12 km and 36 km along the channel axis (Figs. 11a
channel belt in this segment, some of them are located on the and 11c). This relationship is consistent with previous studies
inner banks (Fig. 10). These scars were formed 70 m to 140 m (Babonneau et al., 2002)
above the present-day channel thalweg, suggesting that they Apart from a widened channel floor, decreased channel

8
 ACCEPTED MANUSCRIPT
gradient could also lead to reduced flow energy, sediment 7.1.2 Valley morphology
deposition, widened channel and decreased channel height (e.g. The valley morphology reveals spatial variations of
Friedmann et al., 2000; Adeogba et al., 2005; Estrada et al., erosional processes within the valley. It is difficult to identify
2005). However, this is not strictly valid for the submarine the erosional mechanism in Segment A. But the morphology
channel in this work. In this example, decreasing channel and architecture of the valley reveal the spatial changes of
gradient from Segment a to Segment b and c is accompanied by erosional processes in Segments B and C.
decreased channel width (Table 1). Similar relationship is also In Segment B, the valley base is more than two times
observed from Segment d to Segment e (Table 1). Additionally, wider than other segments (Table 2). This difference is
increased channel height with decreased channel gradient from interpreted as results of cut bank erosion during multiple lateral

PT
Segment c to e also contradicts previous work. Therefore, channel migration (Fig. 15a), as shown by channel forms at the
channel gradient probably is not the main cause for the valley base (Figs. 13c and 13d). Cantilever failure probably is
variation of channel size in this work. the main erosional processes in this segment.

RI
Because the height of leveed channel indicates the In Segment C, inner bank erosion in the form of
minimum thickness of turbidity-flows forming internal levees, shallow-seated bank failures (Fig. 13e-13g) widened the upper

SC
and thus can be used as an indicator of flow volume bank of the valley, and resulted in stepped bank profile and
(Babonneau et al., 2002; Deptuck et al., 2003; Estrada et al., widened upper bank (Fig. 15b). Similar scars have also been
2005). Spatial variations of channel height along the channel recorded in other submarine channels (e.g. Deptuck et al., 2007;
length suggests that flows flushed the channel were not the Janocko et al., 2013). Additionally, shallow-seated bank

U
same. Such variations may reveal spatial and temporal failures are mainly located in Segment C (Fig. 12), suggesting
AN
variations of flow volume. This differ from the model result that the role of local factors on the bank erosion processes. The
from Sylvester et al. (2011), who found that the valley with marked increase in valley height from Segment B to Segment C
complex architecture can be formed by a single channel size could represent this, as the increase may have been associated
M

though time (Sylvester et al., 2011). with the occurrence of scars and associated mass-wasting in
The relationship between temporal change in flow Segment C (Fig. 11b).
properties and channel size has been recorded in Niger delta, The spatial variations of the valley morphology suggest
D

where temporal decrease in channel size is correlated to the the valley size could change over a short distance (i.e. a few
decrease in sediment supply (Jobe et al., 2015). Similar kilometers), such as twofold increase in valley CSA in Segment
TE

scenario may also occur in the studied channel. The volume of B for 7 km (Table 2). Such an increase in valley size would
turbidity flows flushed the channel may have reduced through increase the volume of subsequent channel-fill deposits and
time, leading to decreased channel size in Segment b, c d, and e. reservoir potential, but the extent of valley enlargement is
EP

Additionally, tributaries in the confluence region could relatively small, which increases the uncertainty of drilling.
contribute to the variations in channel size. Apart from the
junction of the west tributary with the main seafloor channel at 7.2 Scale comparison with other submarine
C

confluence point, two other tributary channels connect to the channel systems
east tributary are observed on the seafloor (Fig. 9). These
AC

tributaries may have provided small-volume, less energetic This work shows a channel of 243-1061 m wide, and 13 –
flows to the main pathway. These flows tended to deposit in 156 m high (Figs. 11b and 11c). Channels hundreds of meters
Segments b and c, rather than transport sediments further wide, tens of meters high have been shown on seismic data,
downslope in Segments d and e, thus created small-sized such as Gulf of Mexico (Sylvester et al., 2012) and Congo
channel in Segments b and c. (Deptuck et al., 2007; Jobe et al., 2015), and on bathymetric
Despite of considerable changes of channel size, the width data from California (Maier et al., 2012). Similar sized
of channel floor shows relatively small changes along the channels have also been documented from outcrop studies
channel (Fig. 11c and Table 2), suggesting the channel floor has (Brunt et al., 2013; Figueiredo et al., 2013; Gardner et al., 2003;
been slightly affected despite of variations in flow discharge. Di Celma et al., 2011; Moody et al., 2012; Bain and Hubbard,
2016). Furthermore, channels of kilometers wide are recorded

9
 ACCEPTED MANUSCRIPT
in Amazon fan and Zaire fan, where channels are wider than 1 are not perpendicular to the channel and valley axis, while they
km (Pirmez and Flood, 1995; Babonneau et al., 2002). As may be underestimated because muddy deposits at the top of
submarine channels are products of turbidity flows, the changes channel fill tend to be eroded and poorly exposed (McHargue et
of their size are probably reflect variations in flows properties al., 2011).
(e.g. flow volume, grain-size distribution) (e.g. Babonneau et
al., 2002; Pirmez and Imran, 2003; Sequeiros, 2012; Konsoer et 8. Conclusions
al., 2013; Jobe et al., 2015).
The valley is 700 – 2800 m wide, 230 – 350 m high in this This study focuses on the geomorphology and bank
study. Similar-sized valleys have been documented from both erosion processes in a late Pleistocene-Quaternary channel

PT
seismic data (e.g. Wood and Mize-spansky, 2009; Gamberi et system developed in the Espírito Santo Basin, SE Brazil, to
al., 2013) and outcrop studies (e.g. Masalimova et al., 2016). conclude:
However, the size of valleys or channel complexes reported in

RI
the literature have shown marked variations. For example, a) On channel scale, five segments are identified based
channel complexes less than 1000 m wide have been recorded on the variations in morphologic parameters. The

SC
from outcrop studies in USA (Pyles et al., 2010; Gardner et al., significant variation in channel morphology may
2003), Italy (Thomas and Bodin, 2013) and Chile (Macauley have related to spatial and temporal changes of flow
and Hubbard, 2013). Meanwhile, valley more than 3000 m volume within the channel.
wide have also been recorded from seismic data (e.g. Samuel et

U
al., 2003; Deptuck et al., 2007; Catteral et al., 2010; De Ruig et b) Valley enlargement was controlled by two distinct
AN
al., 2006; Kolla et al., 2012; Jolly et al., 2016), and outcrop bank erosion processes, as shown by the variation in
studies (e.g. Bain and Hubbard, 2016; Grecula et al., 2003). In valley morphology. Cut bank erosion during lateral
Indus fan, the valley can be up to 10 km (Deptuck et al., 2003). channel migration lead to the retreat of entire banks
M

Such marked variations may have been related to the degree of and valley enlargement, especially widening the
lateral channel migration. This seems to be the case in the valley base in Segment B. In contrast, inner bank
Espírito Santo submarine channel, where the higher degree of erosion in the form of shallow-seated bank failures
D

lateral migration corresponds to higher values of the valley only enlarge the upper part of the bank, as in
width and CSA observed in Segment B. In contrast, the lower Segment C.
TE

degree of lateral migration resulted in a valley with lower width

and CSA. The spatial correlation between channel Segments b c) The scales of channel and valley in this study are
and c, and valley Segment B (Fig. 12) suggests that increased comparable to examples observed in other
EP

degree of lateral channel migration in Segment B may also be continental slopes. Distinct spatial variation in both
associated with local sediment input from tributaries. Besides channel and valley morphology in a same channel
of cut bank erosion during lateral channel migration, inner bank system suggests an important role of local factors
C

erosion has also contributed to the widening process of the (e.g. salt diapirs, tributaries, overbank collapse) on
valley, represented by the shallow-seated bank failure observed the development of the channel system.
AC

in this study.
The different results in channel and valley sizes from Acknowledgements
previous studies are also associated with data source and The authors thank CGG for the permission to use the 3D
measurement methods (Kolla et al., 2001; Wood et al., 2009; seismic data in this article. Stephen Hubbard and an
McHargue et al., 2011). Because of data resolution, seismic anonymous reviewer are acknowledged for their constructive
data is difficult to reveal small sized channel, which may only a comments, which have significantly improved the article.
single wavelength thick. Measurement performed on planform Davide Gamboa publishes with the permission of the Executive
and cross sections could also make a difference in measurement Director, British Geological Survey.
results (e.g. Wood and Mize-spansky, 2009). Furthermore,
channel and valley width may be overestimated when outcrops

10
 ACCEPTED MANUSCRIPT
References Campion, K.M., Sprague, A.R., Mohrig, D., Lovell, R.W., Drzewiecki, P.A.,

Antobreh, A.A., Krastel, S., 2006. Morphology, seismic characteristics and Sullivan, M.D., Ardill, J.A., Jensen, G.N., Sickafoose, D.K., 2000.

development of Cap Timiris Canyon, offshore Mauritania: a newly Outcrop expression of confined channel complexes. In: Weimer, P.,

discovered canyon preserved off a major arid climatic region. Marine Slatt, R.M., Coleman, J., Rosen, N.C., Nelson, H., Bouma, A.H.,

and Petroleum Geology, 23, 37–59. Styzen, M.J., Lawrence, D.T. (Eds.), Deep-Water Reservoirs of the

Abreu, V., Sullivan, M., Pirmez, C., Mohrig, D., 2003. Lateral accretion World. Gulf Coast Section SEPM 20th Bob F. Perkins Research

packages (LAPs): an important reservoir element in deep water sinuous Conference, pp. 127-150.

channels. Marine and Petroleum Geology, 20, 631-648. Catterall, V., Redfern, J., Gawthorpe, R., Hansen, D., Thomas, M., 2010.

Adeogba, A.A., McHargue, T.R., Graham, S.A., 2005. Transient fan Architectural style and quantification of a submarine channel–levee

PT
architecture and depositional controls from near-surface 3-D seismic system located in a structurally complex area: offshore Nile Delta.

data, Niger Delta continental slope. American Association of Petroleum Journal of Sedimentary Research, 80, 991–1017.

Geologists Bulletin, 89, 627-643. Chang, H.K., Kowsmann, R.O., Figueiredo, A.M., Bender, A.A., 1992.

RI
Alves, T.M., Cartwright, J., Davies, R.J., 2009. Faulting of salt-withdrawal Tectonics and stratigraphy of the East Brazil Rift system—an overview.

basins during early halokinesis: effects on the Paleogene Rio Doce Tectonophysics, 213, 97-138.

SC
Canyon system (Espírito Santo Basin, Brazil). American Association of Clark, J.D., Kenyon, N.H., and Pickering, K.T., 1992. Quantitative analysis

Petroleum Geologists Bulletin, 93, 617-652. of the geometry of submarine channels: Implications for the

Alves, T.M., 2012, Scale-relationships and geometry of normal faults classification of submarine fans. Geology, 20, 633–636.

reactivated during gravitational gliding of Albian rafts (Espírito Santo Clark, I.R., Cartwright, J.A., 2009. Interactions between submarine channel

U
Basin, SE Brazil): Earth and Planetary Science Letters, 331-332, 80-96. systems and deformation in deep-water fold belts: examples from the
AN
Bain, H, A., Hubbard, S.M., 2016. Stratigraphic evolution of a long-lived Levant Basin, Eastern Mediterranean Sea. Marine and Petroleum

submarine channel system in the Late Cretaceous Nanaimo Group, Geology, 26, 1465-1482.

British Columbia, Canada. Sedimentary Geology, 337, 113-132. Cobbold, P.R., Meisling, K.E., Mount, V.S., 2001. Reactivation of an
M

Babonneau, N., Savoye, B., Cremer, M., Klein, B., 2002. Morphology and obliquely rifted margin, Campos and Santos basins, southeastern Brazil.

architecture of the present canyon and channel system of the Zaire American Association of Petroleum Geologists Bulletin. 85,

deep-sea fan. Marine and Petroleum Geology, 19, 445–467. 1925-1944.

D

Babonneau, N., Savoye, B., Cremer, M., Bez, M., 2004. Multiple terraces Cordani, U.G., 1970. Idade do vulcanismo no Oceano Atlântico Sul.

within the deep incised Zaire Valley (ZaiAngo Project): are they Boletim do Instituto de Geociências e Astronomia da Universidade de
TE

confined levees?. Geological Society of London, Special Publication, São Paulo, 1, 9-76.

222, 91-114. Davison, I., 1999. Tectonics and hydrocarbon distribution along the

Bastos, C., Quaresma, V., Marangoni, M., D’Agostini, D., Bourguignon, S., Brazilian South Atlantic margin. In: N.R. Cameron, R.H. Bate and V.S.
EP

Cetto, P., Silva, A., Filho, G., Moura, R., Collins, M., 2015. Shelf Clure (Eds.), The oil and gas habitats of the South Atlantic. Geological

morphology as an indicator of sedimentary regimes: A synthesis from a Society London Special Publications. 153, 133-151.

mixed siliciclastic–carbonate shelf on the eastern Brazilian margin. Demercian, S., Szatmari, P., Cobbold, P.R., 1993. Style and pattern of salt
C

Journal of South American Earth Sciences, 63, 125-136. diapirs due to thin-skinned gravitational gliding, Campos and Santos

Baudon, C., Cartwright, J., 2008. The kinematics of reactivation of normal basins, offshore Brazil. Tectonophysics. 228, 393-433.
AC

faults using high resolution throw mapping. Journal of Structural Deptuck, M.E., Steffens, G.S., Barton, M., Pirmez, C., 2003. Architecture

Geology, 30, 1072-1084. and evolution of upper fan channel‐belts on the Niger Delta slope and

Bruhn, C.H.L., Walker, R.G., 1997. Internal architecture and sedimentary in the Arabian Sea. Marine and Petroleum Geology, 20, 649-676.

evolution of coarse‐grained, turbidite channel‐levee complexes, early Deptuck, M.E., Sylvester, Z., Pirmez, C., O'Byrne, C., 2007. Migration‐

eocene Regência canyon, Espirito Santo Basin, Brazil. Sedimentology, aggradation history and 3‐D seismic geomorphology of submarine

44, 17-46. channels in the Pleistocene Benin‐major Canyon, western Niger Delta

Brunt, R., Di Celma, C., Hodgson, D., Flint, S., Kavanagh, J., Van der slope. Marine and Petroleum Geology, 24, 406-433.

Merwe, W., 2013. Driving a channel through a levee when the levee is De Ruig, M.J., Hubbard, S.M., 2006. Seismic facies and reservoir

high: an outcrop example of submarine down-dip entrenchment. characteristics of a deep-marine channel belt in the Molasse foreland

Marine and Petroleum Geology, 41, 134-145. basin, Puchkirchen Formation, Austria. American Association of

11
 ACCEPTED MANUSCRIPT
Petroleum Geologists Bulletin, 90, 735–752. slope to basin profile, Brushy Canyon Formation, West Texas. In:

Di Celma, C.N., Brunt, R.L., Hodgson, D.M., Flint, S.S., Kavanagh, J.P., Bouma, A.H., Stone, C.G. (Eds.), Fine-Grained Turbidite Systems.

2011. Spatial and temporal evolution of a Permian submarine slope Memoir 72-American Association of Petroleum Geologists and Special

channel-levee system, Karoo Basin, South Africa. Journal of Publication 68-SEPM, pp. 195-214.

Sedimentary Research, 81, 579–599. Gardner, M.H., Borer, J.M., Melik, J.J., Mavilla, N., Dechesne, M., Wagerle,

Dominguez, J.M.L., Bittencourt, A.C.S.P., Martin, L., 1992. Controls on R.D., 2003. Stratigraphic process-response model for submarine

Quaternary coastal evolution of the east-northeastern coast of Brazil: channels and related features from studies of Permian Brushy Canyon

roles of sea-level history, trade winds and climate. Sedimentary outcrops, West Texas. Marine and Petroleum Geology, 20, 757–788.

Geology, 80, 213-232. Gee, M.J.R., Gawthorpe, R.L., Bakke, K., Friedmann, S.J., 2007, Seismic

PT
Estrada, F., Ercilla, G., Alonso, B., 2005. Quantitative study of a Magdalena geomorphology and evolution of submarine channels from the Angolan

submarine channel (Caribbean Sea): implications for sedimentary continental margin. Journal of Sedimentary Research, 77, 433–446.

dynamics. Marine and Petroleum Geology, 22, 623–635. Georgiopoulou, A., Cartwright, J.A., 2013. A critical test of the concept of

RI
Fiduk, J.C., Brush, E.R., Anderson, L.E., Gibbs, P.B., Rowan, M.G., 2004. submarine equilibrium profile. Marine and Petroleum Geology, 41,

Salt deformation, magmatism, and hydrocarbon prospectivity in the 35-47.

SC
Espirito Santo Basin, offshore Brazil. In: Post, P.J., et al. (Eds.), Grecula, M., Flint, S.S., Wickens, H.D., Johnson, S.D., 2003.

Salt-sediment Interactions and Hydrocarbon Prospectivity: Concepts, Upward-thickening patterns and lateral continuity of Permian sand-rich
st
Applications, and Case Studies for the 21 Century. GCSSEPM 24th turbidite channel fills, Laingsburg–Karoo, South Africa. Sedimentology,

Annual Research Conference, pp. 370-392. 50, 831–853.

U
Figueiredo, J.J.P., Hodgson, D.M., Flint, S.S., Kavanagh, J.P., 2013. Hansen, L.A.S., Callow, R.H.T., Kane, I.A., Gamberi, F., Rovere, M.,
AN
Architecture of a channel complex formed and filled during long-term Cronin, B.T., Kneller, B.C., 2015. Genesis and character of thin-bedded

degradation and entrenchment on the upper submarine slope, Unit F, turbidites associated with submarine channels. Marine and Petroleum

Fort Brown Fm., SW Karoo Basin, South Africa. Marine and Geology, 67, 852–879.
M

Petroleum Geology, 41, 104–116. Hubbard, S.M., Covault, J.A., Fildani, A., Romans, B.R., 2014. Sediment

Flood, R.D., Damuth, J.E., 1987. Quantitative characteristics of sinuous transfer and deposition in slope channels: deciphering the record of

distributary channels on the Amazon deep-sea fan. Geological Society enigmatic deep-sea processes from outcrop. Geological Society of
D

of America Bulletin, 98, 728-738. America Bulletin, 126, 857–871.

França, R.L., Del Rey, A.C., Tagliari, C.V., Brandão, J.R., De Rossi Hubbard, S.M., De Ruig, M.J., and Graham, S.M., 2009. Confined
TE

Fontanelli, P., 2007. Bacia do Espírito Santo. Boletim de Geociencias channel-levee complex development in an elongate depocenter:

da Petrobras, 15, 501-509. Deep-water Tertiary strata of the Austrian Molasse Basin. Marine and

Friedmann, S.J., Beaubouef, R.T., Pirmez, C., Jennette, D.C., 2000. The Petroleum Geology, 26, 85–112.
EP

effects of gradient changes on deep-water depositional systems: an Hubbard, S.M., Romans, B.W., Graham, S.A., 2008. Deep-water foreland

integrated approach: American Association of Petroleum Geologists, basin deposits of the Cerro Toro Formation, Magallanes basin, Chile:

2000 Annual Meeting, Extended Abstracts, New Orleans, U.S., pp. 51. architectural elements of a sinuous basin axial channel belt.
C

Gamboa, D., Alves, T. M., 2015. Spatial and dimensional relationships of Sedimentology, 55, 1333–1359.

submarine slope architectural elements: A seismic-scale analysis from Janocko, M., Nemec, W., Henriksen, S., Warchoł, M., 2013. The diversity of
AC

the Espírito Santo Basin (SE Brazil). Marine and Petroleum Geology, deep-water sinuous channel belts and slope valley-fill complexes.

64, 43-57. Marine and Petroleum Geology, 41, 7-34.

Gamboa, D., Alves, T. M., Cartwright, J., 2012. A submarine channel Jobe, Z.R., Sylevster, Z., Parker, A.O., Howes, N., Slowey, N., Pirmez, C.,

confluence classification for topographically confined slopes. Marine 2015. Rapid adjustment of submarine channel architecture to changes

and Petroleum Geology, 35, 176-189. in sediment supply. Journal of Sedimentary Research, 85, 729-753.

Gamberi, F., Rovere, M., Dykstra, M., Kane, I.A., Kneller, B.C., 2013. Jolly, B.A., Lonergan, L., Whittaker, A.C., 2016. Growth history of

Integrating modern seafloor and outcrop data in the analysis of slope fault-related folds and interaction with seabed channels in the toe-thrust

channel architecture and fill. Marine and Petroleum Geology, 41, 83– region of the deep-water Niger delta. Marine and Petroleum Geology,

103. 70, 58-76.

Gardner, M.H., Borer, J.M., 2000. Submarine channel architecture along a Kane, I., Hodgson, D., 2011. Sedimentological criteria to differentiate

12
 ACCEPTED MANUSCRIPT
submarine channel levee subenvironments: exhumed examples from The Lower Mount Messenger Formation, Taranaki Basin, New Zealand.

the Rosario Fm. (Upper cretaceous) of Baja California, Mexico, and Marine and Petroleum Geology, 71, 360-390.

the Fort Brown Fm. (Permian), Karoo Basin, S. Africa. Marine and Mayall, M., Jones, E., Casey, M., 2006. Turbidite channel reservoirs‐Key

Petroleum Geology, 28, 807-823. elements in facies prediction and effective development. Marine and

Kolla, V., Bourges, P., Urrity, J.M., Safa, P., 2001. Evolution of deepwater Petroleum Geology, 23, 821-841.

tertiary sinuous channels offshore, Angola (West Africa) and Mayall, M., Stewart, I., 2000. The architecture of turbidite slope channels.

implications to reservoir architecture. American Association of In: Weimer, P., Slatt, R.M., Coleman, J.L., Rosen, N., Nelson, C.H.,

Petroleum Geologists Bulletin, 85, 1373–1405. Bouma, A.H., Styzen, M., Lawrence, D.T. (Eds.), Global Deep-Water

Kolla, V., 2007. A review of sinuous channel avulsion patterns in some Reservoirs: Gulf Coast Section SEPM Foundation 20th Annual Bob F

PT
major deepsea fans and factors controlling them. Marine and Petroleum Perkins Research Conference, pp. 578-586.

Geology, 24, 450-469. McHargue, T., Pyrez, M.J., Sullivan, M.D., Clark, J.D., Fildani, A., Romans,

Kolla, V., Bandyopadhyay, A., Gupta, P., Mukherjee, B., Ramana, D.V., B.W., Covault, J.A., Levy, M., Posamentier, H.W., Drinkwater, N.J.,

RI
2012. Morphology and internal structure of a recent upper Bengal 2011. Architecture of turbidite channel systems on the continental slope:

Fan-Valley Complex. In: Prather, B.E., Deptuck, M.E., Mohrig, D., Van patterns and predictions. Marine and Petroleum Geology, 28, 728–743.

SC
Hoorn, B., Wynn, R.B. (Eds.), Application of the Principles Seismic Mohriak, W.U., 2003. Bacias sedimentares da margem continental Brasileira.

Geomorphology to Continental Slope and Base-of-slope Systems: Case In: L.A. Bizzi, C. Schobbenhaus, R.M. Vidotti and J.H. Goncalves

Studies from Seafloor and Near-Seafloor. SEPM Special Publication (Eds.), Geologia, Tectonica e Recursos Minerais do Brasil. CPRM,

No vol. 99, pp. 347–369. Brasilia, pp. 87-165.

U
Knoppers, B., Ekau, W., Figueiredo, A.G., 1999. The coast and shelf of east Mohriak, W.U., Nemcok, M., Enciso, G., 2008. South Atlantic divergen
AN
and northeast Brazil and material transport. Geo-Marine Letters, 19, margin evolution: rift-borded uplift and salt tectonics in the basins of

171-178. Southeastern Brazil. In: Pankhurst, R.J., Trouw, R.A.J., Brito Neves,

Konsoer, K., Zinger, J., Parker, G., 2013. Bankfull hydraulic geometry of B.B., de Wit, M.J. (Eds.), West Gondwana Pre-Cenozoic Correlations
M

submarine channels created by turbidity currents: relations between across the South Atlantic Region, Geological Society London, Special

bankfull channel characteristics and formative flow discharge: Journal Publication. 294, 365-398.

of Geophysical Research: Earth Surface, 118, 216-228. Moody, J.D., Pyles, D.R., Clark, J., Bouroullec, R. 2012. Quantitative
D

Lien, T., Walker, R.G., Martinsen, O.J., 2003. Turbidites in the Upper outcrop characterization of an analog to weakly confined submarine

carboniferous Ross Formation, western Ireland: reconstruction of a channel systems: Morillo 1 member, Ainsa Basin, Spain. American
TE

channel and spillover system. Sedimentology, 50, 113–148. Association of Petroleum Geologists Bulletin, 96, 1813-1841.

Love, F., Brush. E., Fiduk, J. C., Gibbs, P., Brink-Larsen, S., Farrow, G., Ojeda, H.A.O., 1982. Structural framework, stratigraphy, and evolution of

2005. Northern Espírito Santo basin canyon models ancient sand Brazilian marginal basins. American Association of Petroleum
EP

transport. Offshore, 74-78. Geologists Bulletin. 66, 732-749.

Macauley, R.V., Hubbard, S.M., 2013. Slope channel sedimentary processes Ortiz-Karpf, A., Hodgson, D.M., McCaffrey, W.D., 2015. The role of

and stratigraphic stacking, Cretaceous Tres Pasos Formation slope mass-transport complexes in controlling channel avulsion and the
C

system, Chilean Patagonia. Marine and Petroleum Geology, 41, 146– subsequent sediment dispersal patterns on an active margin: The

162. Magdalena Fan, offshore Colombia. Marine and Petroleum Geology,

AC

Maier, K.L., Fildani, A., McHargue, T.R., Paull, C.K., Graham, S.A., Caress, 64, 58-75.

D.W., 2012. Punctuated deep-water channel migration: high-resolution Peakall, J., Kane, I.A., Masson, D.G., Keevil, G., McCaffrey, W.D., and

subsurface data from the Lucia Chica channel system, offshore Corney, R., 2012. Global (latitudinal) variation in submarine channel

California, USA. Journal of Sedimentary Research 82, 1–8. sinuosity. Geology, 40, 11–14.

Martin, L., Suguio, K., 1992. Variation of coastal dynamics during the last Pirmez, C., Flood, R. D., 1995. Morphology and structure of Amazon

7000 years recorded in beach-ridge plains associated with river mouths: Channel. In R. D. Flood, D. J. W. Piper, A. Klaus, et al. (Eds.),

example from the central Brazilian coast. Palaeogeography, Proceedings of the Ocean Drilling Program, Initial Reports, Vol. 155,

Palaeoclimatology, Palaeoecology, 99, 119-140. College Station, TX (Ocean Drilling Program), 23-45.

Masalimova. L.U., Lowe. D.R., Sharman. G.R., King. P.R., Arnot. M.J., Pirmez, C., Imran, J., 2003. Reconstruction of turbidity currents in Amazon

2016. Outcrop characterization of a submarine channel-lobe complex: Channel. Marine and Petroleum Geology, 20, 823-849.

13
 ACCEPTED MANUSCRIPT
Posamentier, H., 2003. Depositional elements associated with a basin floor channel system, the Numidian Flysch formation of Sicily: insights

channel-levee system: case study from the Gulf of Mexico. Marine and from a hierarchical approach. Marine and Petroleum Geology, 41,

Petroleum Geology, 20, 677-690. 163-185.

Pyles, D.R., Jennette, D.C., Tomasso, M., Beaubouef, R.T., Rossen, C., Vigorito, M., Murru, M., Simone, L., 2006. Architectural patterns in a

2010. Concepts learned from a 3D outcrop of a sinuous slope channel multistorey mixed carbonate-siliciclastic submarine channel, Porto

complex: Beacon Channel complex, Brushy Canyon Formation, West Torres Basin, Miocene, Sardinia, Italy. Sedimentary Geology, 186,

Texas, U.S.A. Journal of Sedimentary Research, 80, 67–96. 213–236.

Samuel, A., Kneller, B., Raslan, S., Sharp, A., Parsons, C., 2003. Prolific Wonham, J.B., Jayr, S., Mougamba, R., Chuilon, P., 2000. 3D sedimentary

deep-marine slope channels of the Nile Delta, Egypt. American evolution of a canyon fill (lower Miocene-age) from the Mandorove

PT
Association of Petroleum Geologists Bulletin, 87, 541–560. Formation, offshore Gabon. In: Stow, V.D.A., MayaIl, M. (Eds.),

Sequeiros, O.A., 2012, Estimating turbidity current conditions from channel Deep-Water Sedimentary Systems: New Models for The 21st Century.

morphology: A Froude number approach: Journal of Geophysical Pergamon, Oxford, pp. 175–197.

RI
Research, v. 117, C04003, doi:10.1029 /2011JC007201。 Wood, L.J., Mize-spansky, K.L., 2009, Quantitative seismic geomorphology

Sprague, A.R., Sullivan, M.D., Campion, K.M., Jensen, G.N., Goulding, of a Quaternary leveed-channel system, offshore eastern Trinidad and

SC
D.K., Sickafoose, D.K., Jennette, D.C., 2002. The physical stratigraphy Tobago, northeastern South America. American Association of

of deep-water strata: a hierarchical approach to the analysis of Petroleum Geologists Bulletin, 93, 101–125.

genetically related elements for improved reservoir prediction. Wynn, R.B., Cronin, B.T., Peakall, J., 2007. Sinuous deep‐water channels:

American Association of Petroleum Geologists Annual Meeting Genesis, geometry and architecture. Marine and Petroleum Geology, 24,

U
abstracts, Houston, Texas, pp. 10-13. 341-387.
AN
Sprague, A.R.G., Garfield, T.R., Goulding, F.J., Beaubouef, R.T., Sullivan, Wynn, R.B., Talling, P.J., Masson, D.G., Le Bas, T.P., Cronin, B.T.,

M.D., Rossen, C., Campion, K.M., Sickafoose, D.K., Abreu, V., Stevenson, C.J., 2012. The influence of subtle gradient changes on

Schellpeper, M.E., Jensen, G.N., Jennette, D.C., Pirmez, C., Dixon, deep-water gravity flows: a case study from the Moroccan Turbidite
M

B.T., Ying, D., Ardill, J., Mohrig, D.C., Porter, M.L., Farrell, M.E., system. SEPM Special Publication, 99, 371-383.

Mellere, D., 2005. Integrated slope channel depositional models: the

key to successful prediction of reservoir presence and quality in

D

offshore West Africa. CIPM, cuarto E-Exitep 2005, February 20-23,

2005, Veracruz, Mexico, pp. 1-13.

TE

Stevenson, C.J., Talling, P.J., Wynn, R.B., Masson, D.G., Hunt, J.E., Frenz,

M., Akhmetzhanhov, A., Cronin, B.T., 2013. The flows that left no

trace: very large-volume turbidity currents that bypassed sediment

EP

through submarine channels without eroding the seafloor. Marine and

Petroleum Geology, 41, 186-205.

Sylvester, Z., Pirmez, C., Cantelli, A., 2011. A model of submarine

C

channel-levee evolution based on channel trajectories: implications for

stratigraphic architecture. Marine and Petroleum Geology, 28, 716–

AC

727.

Sylvester, Z., Deptuck, M.E., Prather, B.E., Pirmez, C., O’Byrne, C., 2012.

Seismic stratigraphy of a shelf-edge delta and linked submarine

channels in the North-Eastern Gulf of Mexico, in Prather, B.E.,

Deptuck, M.E., Mohrig, D., van Hoorn, B., and Wynn, R., eds.,

Application of the Principles of Seismic Geomorphology to

Continental-Slope and Base-of-Slope Systems: Case Studies from

Seafloor and Near-Seafloor Analogues: SEPM, Special Publication 99,

p. 31–59.

Thomas, M., Bodin, S., 2013. Architecture and evolution of the Finale

14
 ACCEPTED MANUSCRIPT

PT
RI
SC
Figure 1. (a) Regional map of the SE Brazilian Margin showing the location of the studied 3D seismic volume

U
from the Espírito Santo Basin. (b) Contoured seafloor map of study area generated from the interpreted seismic
AN
volume highlighting the location of the modern channel relative to adjacent salt diapirs. D1 to D6 highlight the

location of six salt diapirs beneath the seafloor.

M
D
TE
C EP
AC

15
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
C EP
AC

Figure 2. Schematic diagrams showing terminology used in the study. (a) and (b) are uninterpreted and
interpreted seismic sections showing architectures of submarine channel system in this work. (c) Channel is
defined as a negative relief on the seafloor. (b) Channel belt is negative relief on the seafloor, which is composed
of channel and terraces. (d) Valley is composed of channel belt and valley fill. It is formed by lateral migration and
vertical aggradation of the channel.

16
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
C EP

Figure 3. Stratigraphic column of the Espírito Santo Basin, showing main tectono-sedimentary stages and
AC

magmatic events (modified from França et al., 2007)

17
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
Figure 4. Simplified regional cross section across the Espírito Santo Basin, showing the major depositional
sequences. The location of the study area is indicated by a black box (Modified from Fiduk et al. 2004 and
Gamboa et al. 2010).
M
D
TE
C EP
AC

18
 ACCEPTED MANUSCRIPT

PT
RI
SC
Figure 5. Selected seismic sections highlighting the morphology of the Rio Doce canyon systems. The location of

U
the seismic sections is shown in Fig. 1b. Four seismic stratigraphic units are identified in the study area based on
AN
França et al. (2007) and Gamboa et al. (2012). Three main incision stages (Eocene, Miocene and

Pliocene-Quaternary) in the Rio Doce Canyon system are highlighted in dark in each seismic section.
M
D
TE
C EP
AC

19
 ACCEPTED MANUSCRIPT

PT
RI
U SC
Figure 6. Measurements taken for the Late Pleistocene-Holocene channel system in this work. (a) and (b) are

uninterpreted and interpreted seismic sections of the channel system. (c) Channel-measurement was taken along
AN
the channel axis, including measurement of depth of channel bank and thalweg, width of channel floor and

channel, cross-sectional area (CSA) of channel. Channel height is the distance between depth of channel bank and
M

thalweg. (d) Valley-measurement was taken along the valley axis, including measurement of depth of valley bank
D

and thalweg, width of valley base and valley, cross-sectional area (CSA) of valley. Valley height is the distance

between depth of channel bank and thalweg.

TE
C EP
AC

20
 ACCEPTED MANUSCRIPT

PT
Figure 7. Selected seismic profiles illustrating different seafloor slope trends in the study area. The location of the

RI
seismic profiles is shown in Fig. 1b. (a) Uninterpreted and (b) interpreted seismic profile showing a decreasing

SC
slope trend in NW-SE direction. Diapir D1 decreases gradient upslope and increases gradient downslope. Diapir

D2 has a minor influence on slope. (c) Uninterpreted and (d) interpreted seismic profiles showing a decreasing

U
slope trends in N-S direction. Diapir D5 reduces gradient upslope and increases gradient downslope.
AN
M
D
TE
C EP
AC

21
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D

Figure 8. Schematic representation of major seafloor geomorphologic features in the study area. The location of
TE

salt diapir D2 is shown in the Figure, although no clear dome is presently observed on the seafloor. The

pre-confluence and confluence regions show a variety of erosional features such as gullies, channels, headwall
EP

scars left by mass-wasting events and irregular depressions. In contrast, the post-confluence region is relatively

smooth with fewer erosion features.

C
AC

22
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP

Figure 9. Dip map of the seafloor, showing the seabed channel comprises west and east tributaries upslope and a
C

post confluence channel downslope. Other small tributaries connect the east tributary are also observed on the

seafloor. Both west and east tributaries change their orientation in pre-confluence region, and were diverted into
AC

the confluence region due to salt diapirs. In the confluence and post-confluence region, the general orientation of

the channel changes to nearly N-S until the southern boundary of the seismic volume is reached.

23
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D

Figure 10. Dip map showing the dominate pathway of the channel system. The dominant pathway is divided into
TE

different segments based on channel (Segments a to e)and valley scales (Segments A to C). Various terraces and

scars are present along the channel system.

C EP
AC

24
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
C EP
AC

Figure 11. Quantitative analysis of the channel. (a) Depth profile of channel thalweg and bank along the channel.

(b) Channel height profile. (c) Width of channel floor and channel. (d) Aspect ratio (width/height) of the channel.

(e) Cross-sectional area (CSA) of the channel.

25
 ACCEPTED MANUSCRIPT

Table 1. The result of morphological analysis along the channel.

PT
RI
U SC
AN
M
D
TE
EP
C
AC

26
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
C EP
AC

Figure 12. Quantitative analysis of the valley. (a) Depth profile of valley and bank along the valley. (b) Valley

height profile. (c) Width of valley base and valley. (d) Aspect ratio (width/height) of the valley. (e) Cross-sectional

area (CSA) of the valley.

Table 2. The result of morphological analysis along the valley.
27
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
EP
C
AC

28
 ACCEPTED MANUSCRIPT

PT
RI
U SC
AN
M
D
TE
C EP
AC

Figure 13. Selected seismic profiles in each segment of the channel system.

29
 ACCEPTED MANUSCRIPT

PT
RI
Figure 14. Comparison of stratal hierarchy between this study and outcrop study from Karoo Basin (De Celma et
al., 2011). The channel corresponds to elementary channel, both of which are fundamental element of hierarchy
framework. The valley corresponds to channel complex, both of which are formed by lateral channel migration

SC
and vertical stacking.

U
AN
M
D
TE

Figure 15. Schematic diagram showing the effects of different bank erosion processes on valley morphology. (a)

Cut bank erosion during lateral channel migration caused the retreat of the entire valley wall, leading to the
EP

widening of the valley, especially the valley base. (b) Shallow-seated bank failures only enlarged the uppermost

half of valley wall.

C
AC

30
 ACCEPTED MANUSCRIPT

Highlights:
The channel and valley show marked variation of morphology in the same channel system.

The variations in channel morphology may have related to spatial and temporal variations in flow
discharge.

PT
The variations in valley morphology is influenced by the erosional processes within the valley.

RI
U SC
AN
M
D
TE
C EP
AC