Origin and preservation conditions of organic matter in

the Mozambique Channel: Evidence for widespread

oxidation processes in the deep-water domains
Martina Torelli, Anne Battani, Daniel Pillot, Eric Kohler, Joel Lopes de
Azevedo, Isabelle Kowalewski, Lucie Pastor, Christophe Brandily, Sabine
Schmidt, Gwenael Jouet, et al.

To cite this version:

Martina Torelli, Anne Battani, Daniel Pillot, Eric Kohler, Joel Lopes de Azevedo, et al.. Ori-
gin and preservation conditions of organic matter in the Mozambique Channel: Evidence for
widespread oxidation processes in the deep-water domains. Marine Geology, 2021, 440, pp.106589.
�10.1016/j.margeo.2021.106589�. �hal-03337053�

HAL Id: hal-03337053

https://hal.science/hal-03337053
Submitted on 21 Apr 2023

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est

archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents
entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non,
lished or not. The documents may come from émanant des établissements d’enseignement et de
teaching and research institutions in France or recherche français ou étrangers, des laboratoires
abroad, or from public or private research centers. publics ou privés.
 1

Marine Geology
October 2021, Volume 440, Pages 106589 (16p.)
Archimer
https://doi.org/10.1016/j.margeo.2021.106589 https://archimer.ifremer.fr
https://archimer.ifremer.fr/doc/00711/82267/

Origin and Preservation Conditions of Organic Matter in the

Mozambique Channel: Evidence for Widespread Oxidation
Processes in the Deep-Water Domains

Torelli Martina 1, *, Battani Anne 1, Pillot Daniel 1, Kohler Eric 1, De Azevedo Joel Lopes 1,
Kowalewski Isabelle 1, Pastor Lucie 2, Brandily Christophe 2, Schmidt Sabine 3, Jouet Gwenael 4,
Deville Eric 1

1 IFP Energies Nouvelles, 1-4 Av. de Bois-Préau, 92852, Rueil-Malmaison Cedex, France
2 IFREMER, Centre, de Bretagne, REM/EEP, Laboratoire Environnement Profond, F-29280 Plouzané,
France
3 UMR5805 EPOC, CNRS, OASU, Université de Bordeaux, 33615 Pessac, France
4 IFREMER, Centre, de Bretagne, REM/GM, Laboratoire Cycles Géochimiques et ressources, F-29280

Plouzané, France

* Corresponding author : Martina Torelli, email address : torellimartina.90@gmail.com

Abstract :

The Mozambique and Madagascar margins present major rivers that are responsible for the discharge of
large amounts of terrestrial organic matter (OM) which can influence carbon cycling in marine
environments. Therefore, the Mozambique channel represents a unique case to study the fate of the
organic carbon in deep-water domains. Using a new and extensive data set of sedimentary OM collected
from sediment traps, seafloor sediments and core sediments, we address the origin of the OM that is
transported and deposited in the Mozambique Channel, its degradation state and preservation conditions.
A Rock-Eval 6 survey allowed us to characterize the origin and amount of OM from shallow to deep-water
turbidite systems, between 500 and 4400 m water depth. Rock-Eval 6 performed on suspended sediments
within particle traps at 47 m above the seabed show that the OM is transported into the deep-water domain
with relatively high TOC (between 1.5 and 2.5%). However, the OM is largely oxidized close to the water-
sediment interface (Oxygen Index >300 mg CO2/g TOC). Seafloor sediments sampled to a maximum
depth of 40 cm show lower TOC values compared to those collected from particle traps suggesting that
the degradation of the OM is mainly active at the water-sediment interface. Small concentrations of OM
are preserved within the recent sediments of the distal area of the Zambezi turbidite system below 2500 m
water depth (TOC < 0.5%). Rock-Eval results show that core sediments from the Majunga slope (NW
margin of Madagascar) and the Zambezi slope (Mozambique margin) contain the highest concentration
of terrestrial OM (TOC between 1 and 2%). However, the OM within core sediments from the deep-water
domain is largely oxidized and degraded, probably due to the conjugate effect of low sediment
accumulation rates (SAR) and high permeabilities of the coarse-grained sediments. Consequently, the
deep-water domain of the Mozambique Channel does not seem to be an important sink of terrestrial OM.
This process is reinforced by important bottom water currents which induce the remobilization and
transport of seafloor sediments that lead to higher oxygen exposure time in the uppermost centimeters of
sediments.

Please note that this is an author-produced PDF of an article accepted for publication following peer review. The definitive
publisher-authenticated version is available on the publisher Web site.
 2

Highlights

► Origin and deposition of the Organic Matter in the main physiographic provinces of the Mozambique
Channel ► Rock-Eval 6 data and δ13C isotopic signature to determine the variability of OM composition
within suspended, seafloor and core sediments ► EDS-SEM analysis, Sediment Accumulation Rates and
Oxygen Exposure Times to assess the preservation condition of the OM in the Mozambique Channel ►
Important bottom currents are prone to maintain higher OET by the remobilization of the uppermost
sediments and its OM ► Small concentration of terrestrial OC (TOC < 0.5%) are preserved in seafloor
sediments of deep-water domain

Keywords : organic matter, transport, preservation, oxidation, Mozambique Channel, Rock- Eval,
Scanning Electron Microscopy

Please note that this is an author-produced PDF of an article accepted for publication following peer review. The definitive
publisher-authenticated version is available on the publisher Web site.
 Journal Pre-proof

1. Introduction

More than 80% of the particulate Organic Matter (OM) is buried on continental and insular
margins (Hedges and Keil, 1995). The burial rate of OM in marine settings affects the
sediment Organic Carbon (OC) budget (Burdige 2007) which is still not well understood and
remains questionable (Gordon and Goñi 2003). It is well known that the burial of OC in
marine sediments is a sink for CO2, possibly greater than that resulting from the alteration of
silicates (Gailladert et al. 1999; Galy et al., 2015). Therefore, quantifications of lateral fluxes
of OC from land to oceans are key controlling factor for CO2 exchange with the atmosphere

of
and the natural carbon cycle (Regnier et al. 2013). In addition, transport conditions of OC to
oceans on the geological time scale determine whether it may be a source or sink to the active

ro
carbon cycle (Bröder et al. 2018).
-p
OC is transported mainly by rivers to the ocean, with a flux estimated between 410 TgC.y-1
and 580 TgC.y-1 (Schlesinger and Melack, 1981; Romankevich et al. 2009; Mignard et al.
re
2017). Most of the OM entering continental shelf areas is rapidly re-mineralized (Hedges
lP

1997; Burdige 2005) but a fraction of it can be preserved and eventually reaches deep-water
domains (Baudin et al. 2020; Mignard et al. 2017). The efficiency of such a process is
controlled by the type of OM (e.g. terrestrial or ―type III‖ vs marine or ―type II‖) which can
na

lead to selective degradation and preservation conditions (Hedges et al. 1988; Stow 2001,
Burdige 2007). It is well known that terrestrial OM which is composed mainly of higher plant
ur

residues, is more resistant to degradation than marine OM which derives from marine
Jo

phytoplankton (Cowie et al. 1992; Burdige 2007). Therefore, higher amounts of terrestrial
OM are more likely to migrate from the shelf to the deep-water domains, mainly transported
by lateral movement of the sediments (e.g. turbidites). In contrast, marine OM is generally
dominated by vertical movement of particles (e.g. Mollenhauer et al. 2007). In addition to the
lability of the OM, there are several factors that control its preservation before and after
sedimentation along the continental shelf such as (a) primary productivity and rate of OM
supply (Pedersen and Calvert 1990; Jahnke 1990), (b) oxygen conditions of the water column
and in the uppermost centimeters of sediments (Hartnett et al. 1998), (c) sedimentation rate
(Muller and Suess 1979, Burdige 2007) and, (d) grain size (Keil et al. 1994).

Conventional hypotheses concerning OM preservation did not consider scenarios with large
accumulations of OM in deep-water areas (Suess 1980). However, deep-water depositional

3
 Journal Pre-proof

environments can be characterized by important OM accumulations, as observed by Stow et

al. (2001) in the Makkassar Strait (Indonesia). The terrestrial OM found in Quaternary
sediments is transported through the Mahakam Delta Province by turbidity currents, and
deposited at more than 2000 m of water depth with relatively high Total Organic Carbon
(TOC) content (between 1% and 2%) (Stow et al. 2001). Another example of such
depositional environment is the Congo system in which the terrestrial OM is transported 760
km offshore through the canyon to the distal fan (Vangriesheim et al., 2009; Baudin et al.,
2010; Baudin et al., 2017). Within the distal lobes, TOC values range from 3.5% to 4%
(Baudin et al. 2017) due to an efficient OM preservation process. Also there is a direct
connection between canyon and river. Therefore, the deep-water domain of the Congo system

of
is a great sink of continental OC (Baudin et al. 2017). Maier et al. (2019) discussed the fate of
the OC reaching the deep-water domain along the Monterey Canyon (offshore California) at

ro
water depths between 278 and 1849 m, based on the analysis of OM collected in sediment
-p
traps and sediment cores. According to the isotopic signature of OC from sediment traps, the
water column is more enriched in marine OM than sediments at the sea floor that show a
re
higher terrestrial contribution. The marine OM is better preserved in the water column, as the
effect of sediment transport in the canyon induces deposition and reworking of seafloor
lP

sediments which results in a higher consumption and degradation of the marine OM compared
to the terrestrial ones (Maier et al. 2019).
na

Knowing that, we propose to study the origin, fate and preservation state of OM in the
ur

turbiditic system of the Mozambique Channel to assess whether this area acts as a sink of
carbon as the other systems cited above. This system provides a unique opportunity to study
Jo

the OC in the deep-water domains where major rivers (Fig. 1) are responsible for the
discharge of large amounts of terrestrial OM. We propose a new approach that studies the
distribution of the OM along the entire channel from within the water column to the very
early deposition, using sediment traps, interface cores and piston sediment cores. We combine
Rock-Eval data, elemental analyses of carbon and δ13C values to assess the origin and
degradation-preservation processes during deposition, from shallow to deep-water turbidite
systems (between 500 and 4400 m of water depth). Using additional data of sediment
accumulation rates and oxygen micro-profiles, the evolution of the recently deposited OM is
compared between the main physiographic provinces of the Mozambique Channel. Finally, a
sample from each margin was analyzed using Energy Dispersive X-Ray Spectroscopy –
Scanning Electron Microscopy (EDS-SEM) to identify the nature of the mineralogical

4
 Journal Pre-proof

composition and the lithofacies of the sediments which has an important impact on OM
preservation.

of
ro
-p
re
lP
na
ur
Jo

5
 Journal Pre-proof

2. Geological framework

The Mozambique channel corresponds to a wide straight between Africa and Madagascar
(Fig. 1). It was formed progressively, during more than 140 Ma, since the breakup of
Gondwana (Rabinowitz et al., 1983) and the formation of the oceanic lithosphere of the
Mozambique Basin during Jurassic-Cretaceous times (Mueller and Jokat, 2019; Thompson et
al., 2019). The central part of the Mozambique Channel is still tectonically active today as the
southern prolongation of the eastern branch of East African Rift System, and this tectonic
activity has controlled the development of volcanic systems and related seamounts emerging
locally in Europa and Bassas da India (Deville et al., 2018). This active tectonic setting is also

of
associated with a zone of uplift which has influenced the sedimentary pattern, notably the

ro
Zambezi turbidite system (Fierens et al., 2019). The continental margins of the Mozambique
Channel are mostly steep transform or transtensive margins.
-p
The Mozambique Basin is largely filled by a turbidite system from the Zambezi River, the
re
main river along the east African coast. It is also partly filled by turbidite currents from
Madagascar, mainly focused in the Tsiribihina submarine valley which is sourced mainly by
lP

the Tsiribihina and Mangoki Rivers (Fierens et al., 2019; Fig. 1). The Betsiboka and
Mahavavy Rivers, in the north-western part of Madagascar, are responsible for large
na

sedimentary accumulations on the continental shelf (Berthois and Crosnier, 1966; Wiles et al.,
2009; Pastor et al., 2020; Fig. 1). These rivers coming either from Africa or Madagascar
ur

provide large detrital OM input to the Mozambique Channel.

Jo

In the Mozambique Basin, strong bottom currents have been described even in the very deep-
water areas (between 2000 and 4000 m water depth) (Miramontes et al., 2019). These bottom
currents are responsible for the development of eddies which partly block the Mozambique
undercurrents, enhancing southward flow along the Zambezi Channel. Measured bottom
currents in the Zambezi and Tsiribihina submarine valleys reach more than 50 cm.s−1, and
these strong bottom currents are responsible for strong erosion processes along the Zambezi
deep sea valley (Miramontes et al., 2019).

6
 Journal Pre-proof

3. Material and methods

This study was conducted in the framework of the PAMELA (PAssive Margin Exploration
LAboratory) research project (Bourillet et al., 2013) which included several oceanographic
exploration surveys in the Mozambique Channel, Pamela-Moz01 (Olu, 2014), Pamela-Moz04
(Jouet and Deville, 2015) and Pamela-Moz08 (Khripounoff, 2016). Samples were collected
along the Zambezi system (slope, valley and distal part) and Tsiribihina turbidite systems, as
well as along the continental slope off the Betsiboka and South-Mahavavy Rivers (the largest
rivers of NW Madagascar) (Fig. 1). The main objective of this sampling campaign was to gain
an understanding of the evolution of the OM in Quaternary sediments along the eastern

of
margin of Africa.

ro
Sediment traps in moorings located 47 m above the seabed are described in Miramontes et al.
(2019) and Pastor et al (2020). We will refer to such samples as ―suspended sediments‖.
-p
Recovered suspended sediments were kept at 4°C for treatment in the laboratory. Interface
re
cores were collected with an interface multicorer (MTB, 30-40 cm maximum penetration),
referred as ―seafloor sediments‖. These cores were sliced vertically on-board and samples
lP

were kept frozen (-20°C) to be analyzed in the laboratory. Piston sediment cores were
collected with the Küllenberg corer (KS or KSF, recovered core length up to 12 m) of the R/V
na

Atalante (Pamela-Moz01 survey) and the Calypso corer (CS or CSF, recovered core length up
to 33 m) of the R/V Pourquoi Pas? (Pamela-Moz04 survey). We will refer to such samples as
ur

―core sediments‖. Samples were taken on-board and kept at 4°C to be analyzed in the
laboratory. Sampling systems with sediment traps, interface cores and gravity/piston cores are
Jo

described in Supplementary Material (Fig. S1, Appendix A).

3.1 Rock-Eval
Rock-Eval pyrolysis is a standard method for the study of organic-rich rocks. The method was
originally developed for petroleum applications as it provides a rapid determination of the
content, type and maturity of organic matter (Espitalié et al., 1977, 1984; Lafargue et al.,
1998; Behar et al., 2001). In addition, Rock-Eval also yields information about type and
amount of carbonates within sedimentary rocks (Pillot et al., 2014). Rock-Eval analysis can
be performed either on the bulk rock (Espitalie et al., 1977) or on concentrated organic matter
samples (Behar et al., 2001). Rock-Eval 6 analysis (Lafargue et al., 1998) consists of
pyrolysis during which samples are heated first in a pyrolysis furnace under inert gas, and

7
 Journal Pre-proof

then in a combustion furnace using air which allows to determine the petroleum potential of a
sedimentary rock. The hydrocarbons liberated during the progressive heating are measured
with an FID (Flame Ionization Detector) and form the peak S1 that represents the free thermo-
vaporized hydrocarbons. The peak S2 corresponds to the liquid effluents released during
cracking of organic matter. The residual sediments are then subjected to a progressive
combustion (oxidation under air). CO and CO2 are released during pyrolysis and combustion
and are measured continuously using Infra-Red cells. The peak S3 corresponds to the CO2
content released during pyrolysis of the insoluble organic matter (kerogen). These
measurements quantify the inorganic carbon (MINC) and the TOC of the analyzed samples.
The amount of bulk rock required for Rock-Eval analysis ranges from 50 to 70 mg. In this

of
study, we applied a method used at IFP Énergies Nouvelles (IFPEN) adapted for recent
sediments (Deroo et al., 1983; Deville et al., 2015) which applies a lower heating program

ro
than used normally for the study of consolidated sediments, with an initial plateau of 200°C
-p
(procedure adapted for immature sediments) instead of 300°C. In addition, samples were first
rinsed with fresh water because they were poorly compacted and rich in seawater. Samples
re
were then centrifuged and dried prior to the analysis. These preparation steps were carried out
to overcome problems related to the presence of sodium chloride which can generate HCl and
lP

induce an early destabilization of carbonates during pyrolysis and subsequent damage to the
analyzing device. A similar method has also been used by Baudin et al. (2017). Rock-Eval
na

Total Carbon (TC%) values of all core samples have been cross-checked with TC (%) values
obtained by elemental analyses (Fig. 2A). Also, calcite content (%) obtained with Rock-Eval
ur

(Pillot et al., 2015) has been compared with XRF (X-Ray Fluorescence) data using the XRF
Avaatech scanner (Fig. 2B).
Jo

3.2 OC content and δ13C of sedimentary organic carbon

For specific seafloor sediments (MOZ4-MTB4 and MTB5, and MOZ1-MTB10, MTB12 and
MTB15), Total Carbon (TC %) content was determined using a Leco CNS-2000 auto-
analyzer using acetanilide for calibration. The relative standard deviation (RSD) was typically
< 3% on samples duplicates. Carbonates were removed with HNO3 (30%) at a constant
temperature of 60°C overnight. Subsequently, OC (%) content was determined using the Leco
CS-125 elemental analyzer using steel chips from LECO for the calibration. The RSD of the
measurement was <1% based on 17 standards replicates and typically < 5% on samples
duplicates. Differences between TC (%) and OC (%) allowed to determine the Inorganic
Carbon (MINC) content (%) (Pastor et al. 2020).

8
 Journal Pre-proof

Isotopic signature of the sedimentary organic carbon (δ13C) for 52 samples from specific
sediment traps (i.e. MLP 3, MLP5, MLP7, MLP8) was assessed through the Combustion
Module-Cavity Ring Down Spectroscopy (CM-CRDS - Picarro) (Balslev-Clausen et al.,
2013). Samples were acidified with HCl 1 N at 40 °C until carbonates were completely
dissolved. Samples were rinsed again using milli-Q water to remove acid residues. After a
freeze-drying process, around 10 mg were then analyzed in a small capsule of 5 × 9 mm.
Calibration was performed according to the International Atomic Energy Agency (IAEA)
reference material (Pastor et al. 2020). Precision was typically within 0.03‰ for a triplicate.

3.3 EDS-SEM analysis

of
EDS-SEM techniques are used to determine the microscopic structure of minerals and organic

ro
matter assemblages producing a surface map of the chemical composition of the sample. The
sample is then converted into a mineralogical map of the total matrix. We present here EDS-
-p
SEM results from two samples that are not consolidated and characterized by very low
permeability. The first sample was collected along the Zambezi continental slope at 78 cm bsf
re
from the top of core MOZ04-CS17 (602 m of water depth). The second sample is from the
lP

Majunga continental slope at 54 cm bsf from the top of the core MOZ01-KSF12 (735 m of
water depth). Samples were first dried 1 week at 60°C and subsequently frozen at -80°C and
then lyophilized. Lyophilization is a process that removes the water from low permeability
na

material with limited damage to the micro-structure of the sample. Because of their low
permeability, samples have been impregnated with a depressurization cup at – 0.8 bar to
ur

enable injection of the resin in the small sample pores. The impregnated samples were then
sawed and polished for the SEM analysis. Microscopy was performed with a ―high vacuum‖
Jo

conventional SEM (EVO MA10, Zeiss SMT) in IFPEN laboratories. A motorized five axis
stage provided a first visualization of the samples. The electron microscope is equipped with a
tungsten filament at 15 kV and 100 mA and uses a probe current at 700 to 750 pA for EDS
analysis (mineral identification). It is also used with a probe current at 150 pA for secondary
electron (SE) imaging at high spatial resolution to observe the surface topography, and with a
back scattered electron detector (BSE) to obtain images with atomic number contrast. The
calibration of the silicon drift detector of EDS is done on pure cobalt for quantitative analysis
during 10 s at 10 to 15 kcps (kilo-counts per second), with a dead time of about 15%. In some
cases, XRD is not sufficient to identify minerals. In this case, SEM micrography coupled with
EDS analysis can be used to identify and define mineral phases through their chemical
composition. Chemical micro-analysis is standardized on various minerals with a well-known
9
 Journal Pre-proof

chemical composition and controlled by electron microprobe (Reference standards for X-ray
microanalysis n° 7610 from Micro-Analysis Consultant Limited, UK).

3.4 Sediment accumulation rates

Sediment accumulation rates (SARs) are determined with different methods depending on the
210
available samples. SARs were calculated from depth profiles of excess Pb (210Pbxs; T1/2 =
22.3 years) of interface cores from the Madagascar and Mozambique Margins: MOZ1-MTB6
(Fontanier et al., 2018), MOZ1-MTB3, MOZ4-MTB2 and MOZ4-MTB3 (Pastor et al., 2020)
210 226
and MOZ4-MTB4 and MOZ4-MTB5 (this work). Activities of Pb and Ra were
determined on sediment samples by gamma spectrometry using a well-type, high-efficiency

of
gamma detector ―Canberra‖ (Schmidt and De Deckker, 2015). The detector was calibrated
using IAEA reference material. Activities are expressed in mBq g-1 and errors are based on

ro
210
one standard deviation counting statistics. Excess -p Pb was calculated by subtracting the
226 210
activity supported by its parent isotope, Ra from the total Pb activity in the sediment.
Sediment layers were measured downcore until reaching negligible excess values. Mean
re
210
SARs were calculated from the Pbxs profiles assuming constant flux and constant
sedimentation (referred to as the CF:CS model). SAR along the Zambezi deep sea valley has
lP

been deduced from a single radiocarbon dating performed on foraminifera on the interface
core MOZ1-MTB10 at 35 cm below seafloor. As seafloor sediments were not available in the
na

Tsiribihina valley, a minimum SAR was estimated by converting the mean annual mass flux
measured by a moored sediment trap placed 47 m above the sea floor over a period of 13
ur

months (Miramontes et al., 2019) and assuming a density of 1.5 g cm-3 (average gamma
density measured on piston cores).
Jo

3.5 O2 microprofiling
The O2 profiles were measured on interface cores kept in a refrigerated box to reproduce a
temperature close to the in-situ conditions. For each station, the oxygen concentration in the
bottom water was determined by the Winkler titration technique adapted by Aminot and
Kérouel (2004). The surface water was gently bubbled to avoid any stratification during the
profiling. A microsensor multimeter (Unisense S/A, Denmark) with one to two Clark type
oxygen probes Ox-50 (Revsbech, 1989) with a motor controller were used to record depth
microprofiles inside the sediment with a 50 to 500 μm resolution depending on the oxygen
penetration depth. A linear calibration was performed between the oxygen concentration
measured in the overlying water and an oxygen free solution. Between 1 and 12 profiles were

10
 Journal Pre-proof

performed on each core. For each profile, the oxygen penetration depth (OPD) was estimated.
In situ oxygen exposure times (OET) were calculated as the ratio between OPD and SAR
(Hedges and Keil, 1995, Pastor et al. 2011). OET provide an estimate of the time during
which accumulating uppermost sediments are exposed to molecular oxygen. However,
sediments are likely impacted by lateral OET as function of the sediment transport time which
represent a controlling parameter on OM degradation across continental shelfs (Keil et al.
2004; Bröder et al. 2018). A few estimates of lateral OET give a broad range of impact on
overall OET: Bröder et al (2018) concluded that the lateral OET is order of magnitude higher
than the in-situ in Siberian-Artic land ocean while Keil et al., (2004) concluded that lateral
OET could double the overall OET along the Washington margin. Hence, lateral OET are

of
intimately dependent on regional settings. In this study, transport of OM is mainly driven by
turbidity currents, which can transport a particle several hundreds of km away in few years as

ro
showed in the Congo system (Rabouille et al. 2009). Given that and the high calculated in situ
-p
OET in the distal area, we assume that lateral OET contribution is negligible compared to in
situ OET.
re
lP
na
ur
Jo

11
 Journal Pre-proof

4. Results

The characterization of the OM degradation-preservation processes from shallow to deep-

water turbidite systems of the Mozambique Channel is mainly based on Rock-Eval 6 (RE6)
analyses of core sediments, seafloor sediments and suspended sediments. Full results are
attached in Supplementary Material (Table S1, Appendix B). Here, RE6 results are presented
as average values by piston core, interface core and sediment trap (Table 1). The location of
the sampling sites is presented in Figure 1 and detailed locations are indicated in Table 1.

The TC (%) content obtained by RE6 and Leco shows a good correlation (R2= 0.9; Fig. 2A).

of
Also, CaCO3 content determined from RE6 and Ca content determined from XRF show the

ro
same vertical trend (Fig. 2B). Note that samples presented in Figure 2 were selected from the
same levels, but the analysed material is not exactly the same which explains the small
-p
differences. Therefore, the comparison of (1) TC measured by Rock-Eval method with Leco,
and (2) Inorganic Carbon (MINC) measured with XRF demonstrates that the TOC determined
re
from Rock-Eval is reliable and can be considered calibrated for this study.
lP

For suspended sediments, δ13C analysis of OC, in addition to other proxies like RE6
measurements, could help to better assess OM origin. δ13C values in the range between -17 ‰
na

and -23 ‰ are typical for marine phytoplankton, -24 ‰ to – 30 ‰ for C3 plants (typically
mangroves) (Bouillon et al., 2008), and -8 and -15 ‰ for C4 plants (Yeh and Wang 2001).
ur

Therefore, the measured δ13C in this study, which are slightly lighter than -21 ‰, can be
considered either as a marine contribution or a mix between sources.
Jo

4.1 Analyses of suspended sediments

Particles collected by moored sediment traps located at 47 m above the seabed show TOC
values between 1.5% and 3.5%, S2 peaks between 2.3 and 10 mg HC/g of rock (Fig. 3a), HI
values between 180 and 300 mg HC/g TOC, and OI values between 150 and 300 mg CO2/g
TOC (Fig. 3b). Considering a reference HI value of 400 mg HC/g TOC and 100 mg HC/g
TOC for marine and terrestrial OM respectively (Espitalie et al., 1985), we can assume that
the OM analyzed from particle traps is mainly terrestrial (Table 1). Rock-Eval results seem to
be confirmed by the δ13C values of OC from suspended sediments which vary between -21.5
and -22.1 ‰ (Table 2; Fig. 4). These results suggest that continental-derived OM is transiting
along continental slopes, and is transported down to the distal area of the Zambezi turbidite

12
 Journal Pre-proof

system, thereby preserving high TOC and HI values (Fig. 3). However, the isotopic signature
can also indicate a mixture of marine and terrestrial OM, with a higher input of terrestrial
material. Note that suspended sediments from the distal valley show S2, TOC, HI and OI
values similar to the suspended sediments collected in the Tsiribihina submarine valley or
even in the Majunga margin, rather than the Zambezi valley (Fig. 3b; Table S1).

4.2 Analyses of seafloor sediments

Results obtained from seafloor sediment samples at a maximum depth of 30 – 40 cm (MTB)
show lower values in TOC (0.6 to 1.43%), S2 (2.5 to 10 mg HC/g rock), and HI (180 to 320
mg HC/ g TOC) as compared to suspended sediments (Table 1). OI values are lower than 300

of
mg CO2/g TOC in the Majunga area, in the Tsiribihina channel area and in the distal channel
of the Zambezi turbidite system and up to 600 mg CO2/g TOC along the Zambezi slope

ro
(Table 1 and Supplementary material, Table S1, Appendix B). Tmax values are relatively high
for this completely immature OM which can be interpreted as an advanced state of OM
-p
degradation of humic substances (Hare et al., 2014). Rock-Eval analysis on seafloor
re
sediments were performed on two samples collected at a specific depth interval for each MTB
(0.5 cm and 6.5 cm) (Table S1). TOC value strongly decreases in seafloor sediments located
lP

in the distal depositional areas (Fig. 1, Table S1). Therefore, additional analyses through Leco
were performed on 65 samples from specific MTBs (MOZ1-MTB10, MOZ1-MTB12, MOZ1-
na

MTB15) (Fig. 1) to better determine their OC (%) content (Fig. 5). They yield very low OC
(%) content between 0.2 to 0.6 %. Suspended sediments in the same distal regions show
ur

higher TOC values (between 1.5% to 2.6%) (Table 1; Fig. 3) suggesting that the OM is
transiting to the deep-water domains while minimally affected by degradation. However, the
Jo

OM is rapidly degraded once it is deposited (Fig. 5).

4.3 Analysis of core sediments

Core sediments in the Mozambique Channel have been collected at a maximum depth from
12 to 33 m. They show moderate TOC (TOC < 1.8%), S2 (< 2 mg HC/g of rock) and Tmax
values characteristic of immature sediments (< 420°C) (Table 1). A higher OM content is
observed in the northwestern part of Madagascar (Majunga margin), where the slope is fed by
major rivers such as Betsiboka and South-Mahavavy. Higher OM content is also observed in
the Davie Ridge area and along the Zambezi continental slope (Fig. 6). All samples show a
low HI of around 300 mg HC/g TOC and a higher OI, with a mean value of 300 mg CO2/g
TOC (Table 1). Note that HI and OI values are not given for TOC values < 0.5% because they
13
 Journal Pre-proof

are considered not reliable. These results are characteristic of a terrestrial OM composed
mainly of higher plants (type III) (van Krevelen, 1950). However, we do observe differences
in terms of HI for samples from the Majunga and the Davie Ridge sectors compared to the
Zambezi slope (Fig. 6). Samples show a first trend of HI values between 50 and 100 mg HC/g
TOC, with a mean value of 75 mg HC/g TOC. A second trend is characterized by a mean HI
value of 150 mg HC/g TOC. Therefore, differences in terms of HI may be related to different
oxidation conditions of the initial terrestrial OM, or to a mixture of marine and terrestrial OM.
Sediments from the Zambezi continental slope contain type III OM, with S2 values around 1
mg HC/g rock and a TOC content of up to 1.8%. Sediments collected from the Zambezi
turbidite system, both from the valley and from the distal depositional area, are characterized

of
by low S2 values < 0.5 mg HC/g rock and TOC < 0.8% (Fig. 6). Sediments from the
Tsiribihina turbidite system (sourced from Madagascar) show even lower TOC values

ro
between 0.13 and 0.61%. Finally, samples collected in the distal area of the Mozambique
-p
basin show the lowest S2 (< 0.3 mg HC/g rock) and TOC values (< 0.2%; Fig.6). As such, a
strong OM decrease is observed with increasing water depth, from the Zambezi continental
re
slope to the distal valley of the Zambezi turbidite system (Fig. 7).
lP

According to RE6, following the method of Pillot et al. (2014) which allows us to characterize
and quantify carbonates in solid samples, mineral carbon is present in all samples, mainly in
na

the form of calcium carbonate. Fig. 8 and Fig. 9 show the content of organic carbon (TOC %)
and inorganic carbon (MINC %) in the analyzed core samples as a function of depth.
ur

Depending on the core location, the OM content is either correlated (CS17, CS21, CS25,
CS26) or anti-correlated (CS22; Fig. 9) with the mineral carbon content. In the latter case, the
Jo

dilution of the OM is related to an increase of carbonate-rich turbidites that are sourced from
Bassas da India Island (Fig. 1).

4.4 EDS-SEM analyses on core sediments

EDS-SEM analyses have shown that core sediments from the Zambezi continental slope and
in the Majunga margin (respectively samples MOZ04-CS17-S1-78 and MOZ01-KSF12-S1-
54) are mainly characterized by a matrix composed of very fine sediments and shell debris
principally made of calcite, with a mean size of 100 μm dispersed in the matrix (Fig. 10;
Table 3). All samples show quartz, calcite and feldspar, and minerals are usually well
preserved and not altered. However, a drastic textural difference appears between the
sediments collected from the Zambezi slope and those from the Majunga slope. The samples

14
 Journal Pre-proof

from the Majunga slope show a better preservation compared to those from the Zambezi slope
which appears largely damaged (Fig. 10). The composition and quantities of chemical
elements are very different between these two areas (Table 3). EDS-SEM shows that CS17-
S1-78 is composed mainly of K-feldspar and phyllosilicates such as mica (biotite and
muscovite; Fig. 11). Aggregates of iron oxides are found inside marine bioclasts and inside
OM which is also partly calcified, with accumulation of silica (Fig. 12). The MOZ01-KSF12-
S1-54 sample (offshore Majunga) is composed mainly of calcite and aragonite, with small
amounts of K-feldspar and mica. These sediments are very rich in bioclasts but they do not
show accumulation of iron oxides (Fig. 13) which is in contrast with measurements presented
by Pastor et al. (2020) showing that iron oxides are abundant.

of
4.5 Sediment accumulation rates

ro
210
Pb-derived SARs range from 0.4 to 2.49 cm yr-1 in the Madagascar margin (Fontanier et al.,
-p
2018; Pastor et al., 2020) and from 0.064 to 1.43 cm yr-1 in the Mozambique margin (this
study; Table 4; Fig. S2, Appendix A). In the Zambezi channel, radiocarbon dating gave a
re
14
calibrated age of 7090 ± 150 Cal yrs BP (calibrated years before the present) (6580 ± 30 C
yrs BP) and thus a minimum SAR of approximately 0.005 cm yr-1 (Table 4). In the Tsiribihina
lP

valley, mass accumulation rates in a sediment trap gave a value of 8 mg cm-2 yr-1. This is
equivalent to a sediment accumulation rate of 0.012 cm yr-1 (assuming a density of 1.5 g cm-3
na

for the sediments; Table 4). Note that this SAR value is a minimum estimate because it does
not integrate the sediment supply which occurs laterally between the sediment traps (47 m
ur

above seabed) and the seabed, notably coarse clastic material transported by
Jo

turbidite/contourite currents.

4.6 Oxygen penetration depths

At most sites of the study area, oxygen concentration disappears within the upper 3 cm of
sediment, except in the Tsiribihina (MOZ1-MTB12) and Zambezi valleys (MOZ1-MTB10
and MOZ1-MTB15; Fig. 14) of the deep-water area of the Mozambique Channel. The lowest
measured oxygen penetration depth (OPD) gave a value of 5 mm (Table 4). They were found
in the Madagascar margin in reduced sediments that were affected by methane diffusion
(MOZ4-MTB1 and MOZ4-MTB3; Pastor et al., 2020). The highest OPD were found in the
Tsiribihina (> 9 cm at MOZ1-MTB12) and Zambezi valleys (8.3 cm at MOZ1-MTB10; > 9
cm at MOZ1-MTB15). At the greatest depth which is in the distal part of Zambezi valley
(4070 m at MOZ1-MTB15), concentrations were still above 25 µM at 8 cm (Fig. 14).

15
 Journal Pre-proof

Assuming a continual decrease in oxygen concentration with depth, the OPD in the
Tsiribihina valley and the Zambezi distal channel can be estimated to 11 and 12 cm
respectively (Table 4). OET values are low in the margins (less than 200 years) whereas they
reach several thousands of years in the deep-water area of the Mozambique basin (Table 4).

of
ro
-p
re
lP
na
ur
Jo

16
 Journal Pre-proof

5. Discussion

5.1 Origin of the sedimentary OM

It is generally assumed that carbon flux decreases with water depth due to degradation of OM
within the water column (Suess, 1980). Yet this process holds more for the marine OM flux
(type II) compared to the continental OM flux (type III). Several studies have shown that
terrestrial OM can be transported by down-slope movements and preserved in deep-water
domains (Stow et al., 2001; Mollenhauer et al. 2007; Baudin et al., 2017). In this study we
tried to understand the origin and the fate of the OM transported from the shallow to the deep-

of
water domain of the Mozambique Channel.

ro
RE6 data indicate that suspended sediments from the Zambezi slope are characterized by
-p
higher OI values (300 to 700 mg CO2/g TOC), i.e. more oxidized samples, compared to
suspended sediments from the deep-water Zambezi distal valley (150 to 300 mg CO2/g TOC)
re
(Fig. 3b). The Zambezi distal valley presents suspended sediments which are less oxidized
compared to the Zambezi slope (Fig. 3b), and that show more affinity in terms of S2, TOC, HI
lP

and OI values with suspended sediments collected in the Tsiribihina submarine valley or even
in the Majunga margin (Fig. 3b). According to these observations, the Zambezi distal valley
na

would be preferentially fed by the Madagascar river system rather than from the Mozambique
margin. This interpretation is consistent with the recent uplift caused by active tectonics
ur

across the Mozambique Channel (Deville et al., 2018) and that resulted in the partial closure
of the main Zambezi turbidite channel near the western Bassas da India-Europa area (Fierens
Jo

et al., 2019) (Fig. 1).

According to HI and OI values by Rock-Eval analysis, all the suspended sediments collected
in the main physiographic provinces of the Mozambique Channel seem to be characterized by
terrestrial OM. Based on average values of δ13C between -21.5 and -22.1 ‰, a mixture of
different sources (marine and terrestrial; Fig. 4) cannot be excluded. However, in the
Betsiboka estuary (Fig. 1), values in this range were interpreted as a mixture of C3 and C4
derived OM, with negligible algal inputs (Ralison et al., 2008). Similar values were then
found at the top of interface cores on the continental shelf (Pastor et al., 2020). The OM at
seafloor and in core sediments shows characteristic Rock-Eval 6 values (i.e. HI and OI
values) typical of terrestrial-derived OM (Fig. 6; 15). For comparison, the Monterey Canyon

17
 Journal Pre-proof

in offshore central California shows an increase of marine OM influence in suspended

sediments compared to core sediments (Mayer et al. 2019). The authors propose a scenario in
which sediment density flows in the Monterey Canyon induce reworking and deposition of
seafloor sediments, resulting in preferential consumption of marine OM in cores than in
sediment trap compare to terrestrial OM. In the Mozambique Channel, the presence of
turbidity currents in deep-water domains (Miramontes et al. 2019) can be regarded as
responsible of low organic carbon burial efficiency (Mayer et al. 2019) which would explain a
better preservation of marine OM in sediment traps and δ13C values of mixed OM sources.

5.3 Preservation condition of the sedimentary OM

of
Sediment cores show that the OM is characterized by two HI groups: a first group with

ro
average HI values around 150 mg HC/g TOC (NW of Madagascar and in the Davie Ridge
area), and a second HI group with average values around 75 mg HC/g TOC (Zambezi system)
-p
(Fig. 6). Considering that the suspended sediments collected in the Zambezi slope are already
largely oxidized before deposition (Fig. 3b), we can assume that the different HI groups
re
observed in the sediment cores (Fig. 6) are likely due to different oxidation conditions of an
lP

initial terrestrial OM, as an increase of OM oxidation can lead to higher OI and lower HI
values (Baudin et al. 2017). Additional studies on palynofacies, carbon stable isotopes (δ13C)
of seafloor samples or organic petrography (e.g. optical analysis) could help to confirm this
na

interpretation. We also compare RE6 results of suspended sediments, seafloor sediments and
sediment cores to better assess preservation and transport conditions (Fig. 15). Suspended
ur

sediments contain the highest OM content (TOC values between 1.5 and 3.5%) both in the
Jo

shallow water and in the deep-water domains (Fig. 3a; 15a). The OM seems to be transported
over large distances and minimally affected by degradation processes. Along the Zambezi
turbidite system, OM collected in sediment traps shows that it is preserved with higher TOC
and HI and lower OI values (HI > 200 mg HC/g TOC, OI < 300 mg CO2/g TOC) and
transported by turbidite and/or bottom currents towards the deep-water domains (Fig. 15).
However, below 2500 m of water depth, the OM content in seafloor sediments strongly
decreases, especially in the distal depositional areas (Fig. 16). The decrease of TOC in
seafloor sediments of the deep-water domain of the Mozambique Channel (Fig. 5) is
particularly notable when compared with other turbidite systems such as the Orinoco river
(Deville et al., 2015). The OM signature observed in the Orinoco abyssal plain (2000 to 4000
m of water depth) is similar to the Mozambique Channel (Fig. 3). However, the OM is better
preserved in the recent sediments of the Orinoco abyssal plain which show TOC values
18
 Journal Pre-proof

between 0.5% and 1.5%. We can also compare our findings with the Congo deep-sea fan
where the OM is transported by turbidity currents to 5000 m of water-depth while preserving
high values of TOC (~ 3.5 to 4%) (Baudin et al., 2017) (Fig. 3). Although the OM is well
transported to the distal lobe of the Mozambique Channel as shown by suspended sediments
(TOC between 1.5 and 2.6 %), it is strongly degraded at the water-sediment interface (Fig. 5).
Therefore, we observe that the terrestrial OM sourced by the major rivers along the
Mozambique and Madagascar margins reaches the deep-water domain of the Mozambique
Channel but it is not preserved in the recent sediments. In contrast to the Orinoco or Congo
turbidite systems, the Mozambique Channel does not seem to be an important sink for recent
terrestrial OM. This is probably due to the activity of bottom water currents. Miramontes et al.

of
(2019) have shown that the Mozambique Channel can be defined as a mixed turbidite-
contourite system dominated by erosional processes at the seafloor induced by intense bottom

ro
currents. Therefore, the complexity of water circulation in the basin may be responsible for
-p
the remobilization of the OM which is then diluted and degraded (Fig. 5). This is also
confirmed by the higher OPD in the distal depositional area of the Mozambique Channel (e.g.
re
> 9 cm at MOZ1-MTB12) (Fig. 14) compared to the Congo turbidite system where measured
OPD are ˂ 1 cm (Pozzato et al. 2017).
lP

5.2 Evolution of the OM deposition in the physiographic provinces of the Mozambique

na

Channel

The mineralogical differences in samples from the Zambezi and Majunga slopes could
ur

indicate two different depositional environments which may affect preservation conditions
Jo

(Table 2). The samples from the uppermost sediments of the Mozambique slope present
mainly K-feldspar and mica but no aragonite and only small amounts of calcite (Fig. 9; Table
2). This is also characteristic of an oxidizing depositional environment which is consistent
with the presence of iron oxides in both the mineral matrix and the calcified OM (Fig. 10).
Pore-water profiles measured in the same sediment core show ferruginous conditions (excess
in Fe2+) and labile iron oxides are still present at depth at around 1% (Zindorf et al., 2021).
This suggest that geochemical processes within the sedimentary column in this area are
controlled by iron, even in the sulfate reduction and methane migration zones, and could
explain the precipitation of iron oxide inside marine bioclasts. In contrast, the samples from
the uppermost sediments of NW Madagascar are rich in aragonite, calcite and shell residues,
with no or limited iron oxide (Fig. 11, Table 2). This is representative of reducing
depositional environments in the uppermost cm of sediments. As presented in Pastor et al.
19
 Journal Pre-proof

(2020), the Majunga slope is reached by flows of detrital iron oxides carried by the Betsiboka
river. Therefore, the absence of iron oxides in the MOZ01-KSF12 sample, as revealed by
EDS-SEM analysis, may be related to the consumption of iron oxides in reducing conditions
to form pyrite in relation to upward methane-rich fluid migration (Pastor et al., 2020).

The comparison of analytical results obtained from suspended, seafloor and core samples
suggests that OM is largely diluted and degraded in the first few centimeters of the water-
sediment interface in areas of water depth higher than 2500 m (Fig. 15; 16). In these areas,
oxygen was measured at least down to 8 cm in the sediments and OET reaches several
thousand years (Table 3, Fig. 14). Indeed, in the deep-water area of the Mozambique Channel,

of
important oxidation processes occurred in the uppermost sediments as confirmed by the
decrease of the S2 peak (< 0.5 mg HC/g rock) and the TOC content (< 0.5%), coupled with an

ro
increase of OI (> 300 mg CO2/g TOC) (Fig. 15). This is also confirmed by the thicker O2-
bearing zone in the uppermost sediments of the deep-water area of the Mozambique Basin
-p
along the distal Zambezi and Tsirihibina valleys (Fig. 14). These oxidizing conditions are
re
interpreted by the conjugate effects of lower sedimentation rates (Table 3) and higher
porosities of the sand-rich material of the Zambezi distal system (Fierens et al., 2019). This
lP

process is probably the result of the activity of important bottom currents (Miramontes et al.,
2019) which are prone to maintain oxidizing conditions by the remobilization of the
na

uppermost sediments and its OM (Fig. 15). Indeed, current velocities reach up to 50 cm.s-1 in
the Zambezi valley at MPL2 (Fig. 1), 38 cm.s-1 in the Tsiribihina valley at MPL3 (Fig. 1) and
ur

36 cm.s-1 in the Zambezi distal valley at MPL5 (Miramontes et al., 2019). Overall, these
processes are responsible for higher oxygen exposure times ranging from a few years (7 to
Jo

172 yrs in the Mozambique and Madagascar margins) to thousands of years (up to 16,600 yrs
in the Zambezi distal valley; Table 4).

Bottom currents have a major impact on the evolution of TOC (%) in the main physiographic
provinces of the Mozambique Channel (Fig. 17). The OM is transported to the distal valley as
indicated by higher TOC values in suspended sediments (Fig. 17). However, the OM content
rapidly decreases in the first few centimeters of sediments due to deep and cold water currents
(Miramontes et al. 2019) that induce erosion, reworking and deposition of sediments (Stow et
al. 2018). This process is also confirmed by deeper OPD (Fig. 17). In contrast, OM
transported and deposited along the Majunga and Zambezi valleys is better preserved both in
sediment traps and in core samples (Fig. 3a; Fig. 15). The presence of anticyclonic rings along
the western Mozambique margin (Miramontes et al. 2019) can affect deep-water circulation
20
 Journal Pre-proof

with alternating periods of strong bottom currents that result in seafloor erosion, sediment
resuspension and higher oxidation of its OM. This process is observed in suspended
sediments collected along the Zambezi slope (Fig. 3 b) which contains terrestrial OM already
largely oxidized before deposition. In contrast, the Majunga slope is characterized mainly by
surface water circulation resulting in a less disturbed depositional environment, higher OM
content at the seafloor and lower oxygen penetration depth (Fig. 17).

In the Mozambique Channel, important amount of mixed OM (TOC > 2%) is efficiently
transported until the distal depositional areas (> 4000 m water depth). However, the OM is
not well preserved in areas of water depth higher than 2500 m due to the presence of sediment

of
flows and bottom currents (Miramontes et al. 2019) which induce higher OET and low OC
burial efficiency. Therefore, the distal lobes of the Mozambique Channel are not regarded as a

ro
sink of terrestrial OC in the active carbon cycle.
-p
re
lP
na
ur
Jo

21
 Journal Pre-proof

6. Conclusion

In this work, we present the analysis of the organic matter evolution over a wide scale from
shallow areas to deep-water domains of the Mozambique Channel. Analyses of OM collected
from suspended sediments, seafloor sediments and cores (core length < 33 m of depth)
allowed us to assess transport and degradation-preservation conditions. The Mozambique
Channel accommodates large amounts of terrestrial OM issued from the Madagascar and
Mozambique margins. The highest OM content in core sediments has been found in the
Majunga area (TOC ⁓ 1.4%), offshore Madagascar, and on the Zambezi slope (TOC ⁓ 1.1%),
offshore Mozambique. Suspended sediments from the deep-water domain of the Zambezi

of
turbidite system show a relatively high content of OM (TOC ⁓ 2%) which is less oxidized

ro
compared to the Zambezi slope. Rock-Eval 6 results suggest that the deep-water domains of
the Mozambique Channel are preferentially fed with terrestrial OM issued from the
-p
Madagascar river system rather than from Mozambique. We therefore can assume that in the
deeper parts of the Mozambique Channel, the OM is efficiently transported by turbidity
re
and/or contour currents and deposited in the turbidite system of the deep-water area of the
lP

Mozambique Basin while preserving higher amounts of poorly oxidized OM. However, the
deposited OM in seafloor sediments is largely oxidized. This process is probably related to the
conjugate effects of lower sedimentation rates and higher porosity of the sand-rich material
na

present in the Zambezi distal system. Consequently, the Mozambique Channel does not seem
to be an important sink of terrestrial organic matter in contrast to other turbidite systems. The
ur

activity of important marine currents is responsible for erosion, reworking and remobilization
Jo

of sediments and OM within at the sea bottom, leading to higher oxygen exposure time in the
uppermost centimeters of the sediments.

Acknowledgements

This work has been conducted within the framework of the PAMELA (PAssive Margin
Exploration LAboratories) project sponsored by IFREMER and TOTAL in collaboration with
IFP Energies Nouvelles, Université de Bretagne Occidentale, Université de Rennes 1,
Sorbonne Université and CNRS (Bourillet al., 2013). Data acquisition was performed in 2014
during the PAMELA-MOZ01 campaign (Olu, 2014) onboard the R/V L’Atalante, in 2015
during the PAMELA-MOZ04 campaign (Jouet and Deville, 2015) onboard the R/V Pourquoi
Pas?, and during the PAMELA-MOZ08 campaign (Khripounoff, 2017) onboard the R/V

22
 Journal Pre-proof

Antea. We thank captains, officers, crew members and the scientific teams of these cruises for
their technical support. We also thank Sophie Hage and an anonymous reviewer for providing
constructive and useful comments.

Appendix A. Supplementary material

Supplementary Fig. S1. The three different types of sampling devices

Supplementary Fig. S2. 210Pbxs profiles measured on cores MOZ4-MTB4 and MOZ4-MTB5

of
ro
-p
re
lP
na
ur
Jo

23
 Journal Pre-proof

References

Ahuva, A.L., Amos, B., Sass, E., 1993. Late Cretaceous upwelling system along the Southern
Tethys Margin Israel Interrelationship between productivity, bottom water environments,
and organic matter preservation. Paleoceanography 8, 5, 671–690.
Aminot, A., Kérouel, R., 2004. Hydrologie des écosystèmes marins. Paramètres et analyses.
335 p.
Balslev-Clausen, D., Dahl, T.W., Saad, N., Rosing, M.T., 2013. Precise and accurate delta C-
13 analysis of rock samples using Flash Combustion-Cavity Ring Down Laser
Spectroscopy. J. Anal. At. Spectrom. 28 (4), 516–523. doi:10.1039/C2JA30240C.

of
Barber, A., Brandes, J., Leri, A., Lalonde, K., Balind, K., Wirick, S., Wang, J., Gélinas, Y.,
2017. Preservation of organic matter in marine sediments by inner-sphere interactions with

ro
reactive iron. Scientific Reports 7, 366, doi:10.1038/s41598-017-00494-0.
-p
Baudin, F., Disnar, J-R., Martinez, P., Dennielou, B., 2010. Distribution of the organic matter
in the channel-levees systems of the Congo mud-rich deep-sea fan (West Africa).
re
Implication for deep offshore petroleum source rocks and global carbon cycle. Marine and
Petroleum Geology 27, 5, 995–1010. doi:10.1016/j.marpetgeo.2010.02.006.
lP

Baudin, F., Stetten, E., Schnyder, J., Charlier, K., Martinez, P., Dennielou, B., Droz, L., 2017.
Origin and distribution of the organic matter in the distal lobe of the Congo deep-sea fan –
na

A Rock-Eval survey. Deep Sea Research Part II: Topical Studies in Oceanography
142, 75–90. doi:10.1016/j.dsr2.2017.01.008.
ur

Behar, F., Beaumont, V., B. Penteado, H. L. de, 2001. Rock-Eval 6 Technology.

Performances and Developments. Oil & Gas Science and Technology 56 2, 111–134.
Jo

doi:10.2516/ogst:2001013.
Berner, R. A. (1982). Burial of organic carbon and pyrite sulfur in the modern ocean: its
geochemical and environmental significance. Am. J. Sci.;(United States), 282. DOI:
doi:10.2475/ajs.282.4.451
Berthois, L., Crosnier A., 1966. Etude dynamique de la sédimentation au large de l'estuaire de
la Betsiboka. Cahiers O.R.S.T.O.M., sér. Océanogr. IV, 2.
Bouillon, S., Connolly, R.M., and Lee, S.Y. 2008. Organic matter exchange and cycling in
mangrove ecosystems: recent insights from stable isotope studies. Journal of sea research,
59(1-2), 44-58. doi:10.1016/j.seares.2007.05.001.
Bourillet, J.F., Ferry, J.N., Bourges, P., 2013. PAMELA, Passive Margins Exploration
Laboratories. doi:10.18142/236.

24
 Journal Pre-proof

Bröder, L., Tesi, T., Andersson, A., Semiletov, I., and Gustafsson, Ö. 2018. Bounding cross-
shelf transport time and degradation in Siberian-Arctic land-ocean carbon transfer. Nature
communications, 9(1), 1-8. doi:10.1038/s41467-018-03192-1
Burdige, D. J. 2005. Burial of terrestrial organic matter in marine sediments: A re‐
assessment. Global Biogeochemical Cycles, 19(4). doi:10.1029/2004GB002368.
Burdige, D.J., 2007. Preservation of Organic Matter in Marine Sediments: Controls,
Mechanisms, and an Imbalance in Sediment Organic Carbon Budgets? Chemical Reviews
107, 467-458. doi:10.1021/cr050347q.
Cowie, G. L., Hedges, J. I., & Calvert, S. E. 1992. Sources and relative reactivities of amino
acids, neutral sugars, and lignin in an intermittently anoxic marine environment.

of
Geochimica et Cosmochimica Acta, 56(5), 1963-1978. doi:10.1016/0016-7037(92)90323-
B.

ro
Craig, H. (1953). The geochemistry of the stable carbon isotopes. Geochimica et
-p
cosmochimica acta, 3(2-3), 53-92. doi:10.1016/0016-7037(53)90001-5.
Deroo, G., Herbin, J. P., Roucaché, J., 1983. Organic geochemistry of upper Jurassic
re
Cretaceous sediments from site-511, leg-71, Western South-Atlantic. Initial Reports of the
Deep Sea Drilling Project, 71, 1001–1013.
lP

Deville, E., Mascle, A., Callec, Y., Huyghe, P., Lallemant, S., Lerat, O., Mathieu, X., Padron
de Carillo, C., Patriat, M., Pichot, T., Loubrieux, B., Granjeon, D. 2015. Tectonics and
na

sedimentation interactions in the east Caribbean subduction zone. An overview from the
Orinoco delta and the Barbados accretionary prism. Marine and Petroleum Geology
ur

64, 76–103. doi:10.1016/j.marpetgeo.2014.12.015.

Deville, É., Guerlais, S-H., Callec, Y., Griboulard, R., Huyghe, P., Lallemant, S, Mascle, A.,
Jo

Noble, M.,Schmitz, J. and the CARAMBA research team, 2006. Liquefied vs Stratified
Sedimentation Mobilization Processes: Insight from the South of the Barbados
Accretionary Prism.Tectonophysics, 428, 33-47. doi:10.1016/j.tecto.2006.08.011.
Espitalie, J., Laporte, J.L., Madec, M., Marquis, F., Leplat, P., Paulet, J., 1977. Méthode
rapide de caractérisation des roches mères, de leur potentiel pétrolier et de leur degré
d’évolution. Rev. Inst. fr. petr., 32, 1, 23-42. doi:10.2516/ogst:1977002
Espitalié, J., Madec, M., Tissot, B., Mennig, J. J., Leplat, P., 1977. Source Rock
Characterization Method for Petroleum Exploration. In. Offshore Technology Conference.
OTC, 1977/1/1, Offshore Technology Conference, p. 6.
Espitalié, J., Senga Makadi, K., Trichet, J., 1984. Role of the mineral matrix during kerogen
pyrolysis. Organic Geochemistry 6, 365–382. doi:10.1016/0146-6380(84)90059-7.

25
 Journal Pre-proof

Espitalie, J., Deroo, G., Marquis, F., 1985. Rock-Eval pyrolysis and its applications. I Rev.
Inst. fr. petr. 40, 5, 563–579.
Fierens, R., Droz, L., Toucanne, S., Raisson, F., Jouet, G., Babonneau, N., Miramontes, E.,
Landurain, S., Jorry, S., 2019. Late Quaternary geomorphology and sedimentary processes
in the Zambezi turbidite system (Mozambique Channel). Geomorphology 334, 1–28.
doi:10.1016/j.geomorph.2019.02.033.
Fontanier, C., Mamo, B., Toucanne, S., Bayon, G., Schmidt, S., Deflandre, B., Dennielou, B.,
Jouet, G., Garnier, E., Sakai, S., Lamas, R.M., Duros, P., Toyofuku, T., Salé, A., Belleney,
D., Bichon, S., Boissier, A., Chéron, S., Pitel, M., Roubi, A., Rovere, M., Grémare, A.,
Dupré, S., Jorry, S.J., 2018. Are deep-sea ecosystems surrounding Madagascar threatened

of
by land-use or climate change? Deep Sea Research Part I, Oceanographic Research Papers
131, 93-100. doi:10.1016/j.dsr.2017.11.011.

ro
Gaillardet, J., Dupré, B., Louvat, P., & Allegre, C. J. 1999. Global silicate weathering and
-p
CO2 consumption rates deduced from the chemistry of large rivers. Chemical
geology, 159(1-4), 3-30. doi:10.1016/S0009-2541(99)00031-5.
re
Galy, V., Peucker-Ehrenbrink, B., Eglinton, T., 2015. Global carbon export from the
terrestrial biosphere controlled by erosion. Nature 521, 204. doi:10.1038/nature14400.
lP

Gordon, E. S., & Goñi, M. A. 2003. Sources and distribution of terrigenous organic matter
delivered by the Atchafalaya River to sediments in the northern Gulf of
na

Mexico. Geochimica et Cosmochimica Acta, 67(13), 2359-2375. doi:10.1016/S0016-

7037(02)01412-6.
ur

Hare, A.A., Kuzyk, Z.Z.A., Macdonald, R.W., Sanei, H., Barber, D., 2014. Characterization
of sedimentary organic matter in recent marine sediments from Hudson Bay, Canada, by
Jo

Rock-Eval pyrolysis. Organic Geochemistry 68, 52–60.

doi:10.1016/j.orggeochem.2014.01.007.
Hartnett, H. E., Keil, R.G., Hedges, J.I., Devol, A.H., 1998. Influence on oxygen exposure
time on organic carbon preservation in continental margin sediments. Nature 391, 372-374.
doi:10.1038/35351.
Hedges, J. I., Keil, R.G., 1995. Sedimentary organic matter preservation: an assessment and
speculative synthesis. Marine Chemistry 49, 81-115. doi:10.1016/0304-4203(95)00008-F.
Hedges, J.I., Keil, R.G., Benner, R., 1997. What Happens to Terrestrial Organic Matter in the
Ocean? Organic Geochemistry 27, 5/6, 195–212. doi:10.1016/S0146-6380(97)00066-1.

26
 Journal Pre-proof

Hedges; I. J.; Clark, W.A., Come, G.L. 1988. Organic Matter Sources to the Water Column
and surficial sediments of a marine bay. In Limnol. Oceanogr., 1988, pp. 1116–1136.
doi:10.4319/lo.1988.33.5.1116
Henrichs, S. M., Farrington, J. W., 1984. Peru upwelling region sediments near 15S.
Remineralization and accumulation of organic matter. Limnology and Oceanography 29,
1–19. doi:10.4319/lo.1984.29.1.0001.
Jahnke, R. A. 1990. Early diagnesis and recycling of biogenic debris at the seafloor, Santa
Monica Basin, California. J. Mar. Res. 48, 413–436. doi:10.1357/002224090784988773
Jouet, G., Deville, E., 2015. PAMELA-MOZ04 cruise, R/V Pourquoi Pas?
doi:10.17600/15000700.

of
Keil, R. G., Hu, F. S., Tsamakis, E. C. & Hedges, J. I. 1994. Pollen in marine sediments as an
indicator of oxidation of organic matter. Nature 369, 639–641.

ro
Keil, R. G., Dickens, A. F., Arnarson, T., Nunn, B. L., & Devol, A. H. (2004). What is the
-p
oxygen exposure time of laterally transported organic matter along the Washington
margin?. Marine Chemistry, 92(1-4), 157-165. doi:10.1016/j.marchem.2004.06.024.
re
Khripounoff, A., 2017. PAMELA-MOZ08 cruise, R/V Antea. doi:10.17600/17003900.
Kohn, M. J. (2010). Carbon isotope compositions of terrestrial C3 plants as indicators of
lP

(paleo) ecology and (paleo) climate. Proceedings of the National Academy of

Sciences, 107(46), 19691-19695. doi:10.1073/pnas.1004933107.
na

Lafargue, E., Marquis, F., Pillot, D., 1998. Rock-Eval 6 Applications in Hydrocarbon
Exploration, Production, and Soil Contamination Studies. Rev. Inst. Fr. Pét. 53, 4, 421–
ur

437. doi:10.2516/ogst:1998036.
Maier, K. L., Rosenberger, K. J., Paull, C. K., Gwiazda, R., Gales, J., Lorenson, T., ... &
Jo

Cartigny, M. J. 2019. Sediment and organic carbon transport and deposition driven by
internal tides along Monterey Canyon, offshore California. Deep Sea Research Part I:
Oceanographic Research Papers, 153, 103108. doi:10.1016/j.dsr.2019.103108.
Mignard, S. L. A., Mulder, T., Martinez, P., Charlier, K., Rossignol, L., & Garlan, T. (2017).
Deep-sea terrigenous organic carbon transfer and accumulation: Impact of sea-level
variations and sedimentation processes off the Ogooue River (Gabon). Marine and
Petroleum Geology, 85, 35-53. doi:10.1016/j.marpetgeo.2017.04.009
Miramontes, E., Penven, P., Fierens, R., Droz, L., Toucanne, S., Jorry, S.J., 2019. The
influence of bottom currents on the Zambezi Valley morphology (Mozambique Channel,
SW Indian Ocean). In situ current observations and hydrodynamic modelling. Marine
Geology 410, 42–55. doi:10.1016/j.margeo.2019.01.002.

27
 Journal Pre-proof

Mollenhauer, G., Inthorn, M., Vogt, T., Zabel, M., Sinninghe Damsté, J. S., & Eglinton, T. I.
2007. Aging of marine organic matter during cross‐shelf lateral transport in the Benguela
upwelling system revealed by compound‐specific radiocarbon dating. Geochemistry,
Geophysics, Geosystems, 8(9). doi:10.1029/2007GC001603.
Mueller, C. O., Jokat, W., 2019. The initial Gondwana break-up: A synthesis based on new
potential field data of the Africa-Antarctica Corridor. Tectonophysics, 750, 301–328.
doi:10.1016/j.tecto.2018.11.008.
Müller, P. J., Suess, E. 1979. Productivity, sedimentation rate, and sedimentary organic matter
in the oceans. I. Organic carbon preservation. Deep-Sea Res. A 26, 1347–1362.
doi:10.1016/0198-0149(79)90003-7.

of
Nieto-Cid, M., Alvarez-Salgado, A., Pérez, F.F., 2006. Microbial and Photochemical
reactivity of fluorescent dissolved organic matter in a coastal upwelling system. Limnology

ro
and Oceanography 51, 3, 1391–1400. doi:10.2307/3841185
-p
Olu, K, 2014. PAMELA-MOZ01 cruise, R/V L'Atalante, .
Pastor, L., Deflandre, B., Viollier, E., Cathalot, C., Metzger, E., Rabouille, C., Escoubeyrou,
re
K., Lloret, E., Pruski, A.M., Vetion, G., Desmalades, M., Buscail, R., and Gremare, A.
(2011). "Influence of the organic matter composition on benthic oxygen demand in the
lP

Rhone River prodelta (NW Mediterranean Sea)." Continental Shelf Research 31(9): 1008-
1019. doi:10.1016/j.csr.2011.03.007.
na

Pastor, L., Brandily, C., Schmidt, S., Miramontes, E., Péron, M., Apperé, D., Besson, S.,
Boissier, A., Jouet, G., 2020. Modern sedimentation and geochemical imprints in
ur

sediments from the NW Madagascar margin. Marine Geology, 426, 106184.

doi:10.1016/j.margeo.2020.106184
Jo

Pedersen, T.F., Calvert, S.E., 1990. Anoxia vs. Productivity. What Controls the Formation of
Organic-Carbon-Rich Sediments and Sedimentary Rocks?. AAPG Bulletin 74, 4, 454–466.
doi:10.1306/0C9B282B-1710-11D7-8645000102C1865D.
Pelegrí, J.L., Arístegui, J., Cana, L., González-Dávila, M., Hernández-Guerra, A., Hernández-
León, S., 2005. Coupling between the open ocean and the coastal upwelling region off
northwest Africa. Water recirculation and offshore pumping of organic matter. Journal of
Marine Systems 54, 1-4, 3–37. doi:10.1016/j.jmarsys.2004.07.003.
Pillot, D., Deville, E., Prinzhofer, A., 2014. Identification and Quantification of Carbonate
Species Using Rock-Eval Pyrolysis. Oil & Gas Science and Technology 69, 2, 341–349.
doi:10.2516/ogst/2012036.

28
 Journal Pre-proof

Pozzato, L., Cathalot, C., Berrached, C., Toussaint, F., Stetten, E., Caprais, J.C., Pastor, L.,
Olu-Leroy, K., and Rabouille, C. 2017. Early diagenesis in the Congo deep-sea fan
sediments dominated by massive terrigenous deposits: Part I–Oxygen consumption and
organic carbon mineralization using a micro-electrode approach. Deep Sea Research Part
II: Topical Studies in Oceanography, 142, 125-138. doi:10.1016/j.dsr2.2017.05.010.
Rabinowitz, P.D., Coffin, M.F., Falvey, D., 1983. The Separation of Madagascar and Africa.
Science 220, 4592, 67–69. doi:10.1126/science.220.4592.67.
Rabouille, C., Caprais, J. C., Lansard, B., Crassous, P., Dedieu, K., Reyss, J. L., and
Khripounoff, A. 2009. Organic matter budget in the Southeast Atlantic continental margin
close to the Congo Canyon: In situ measurements of sediment oxygen consumption. Deep

of
Sea Research Part II: Topical Studies in Oceanography, 56(23), 2223-2238.
Ralison, O.H., Borges, A.V., Dehairs, F., Middelburg, J.J., and Bouillon, S. (2008). Carbon

ro
biogeochemistry of the Betsiboka estuary
-p (north-western Madagascar). Organic
Geochemistry, 39(12), 1649-1658. doi:10.1016/j.orggeochem.2008.01.010.
Regnier, P., Friedlingstein, P., Ciais, P. et al. 2013. Anthropogenic perturbation of the carbon
re
fluxes from land to ocean. Nature Geosci 6, 597–607. doi:10.1038/ngeo1830.
Revsbech, N.P., 1989. An oxygen microsensor with a guard cathode. Limnology and
lP

Oceanography 34, 2, 474-478. doi:10.4319/lo.1989.34.2.0474

Romankevich, E.A., Vetrov, A.A., Peresypkin, V.I., 2009. Organic matter of the world
na

ocean. Russ. Geol. Geophys. 50, 299e307. doi:10.1016/j.rgg.2009.03.013.

Schlesinger, W.H., Melack, J.M., 1981. Transport of organic carbon in the world's
ur

rivers. Tellus 33, 172e187. doi:10.1111/j.2153-3490.1981.tb01742.x.

Schmidt, S., De Deckker, P., 2015. Present-day sedimentation rates on the southern and
Jo

southeastern Australian continental margins. Australian Journal of Earth Sciences 62, 143-
150. doi:10.1080/08120099.2015.1014846.
Stow, D.A.V., Huc, A. Y., Bertrand, P., 2001. Depositional Processes of Black Shales in
Deep Water. Marine and Petroleum Geology 18, 491–498. doi:10.1016/S0264-
8172(01)00012-5.
Stow, D., Smillie, Z., & Esentia, I. P. (2018). Deep-Sea bottom currents: Their nature and
distribution. In Encyclopedia of ocean sciences: earth systems and environmental sciences.
doi:10.1016/B978-0-12-409548-9.10878-4.
Suess, E., 1980. Particulate organic carbon flux in the oceans—surface productivity and
oxygen utilization. Nature 288, 5788, 260. doi:10.1038/288260a0.

29
 Journal Pre-proof

Thompson, J.O., Moulin, M., Aslanian, D., de Clarens, P., Guillocheau, F., 2019. New
starting point for the Indian Ocean : Second phase of breakup for Gondwana. Earth-
Science Reviews, 191, 26–56. doi:10.1016/j.earscirev.2019.01.018.
Vangriesheim, A., Khripounoff, A., Crassous, P., 2009. Turbidity events observed in situ
along the Congo submarine channel. Deep-Sea Research Part II-Topical Studies in
Oceanography 56, 23, 2208-2222. doi:10.1016/j.dsr2.2009.04.004.
Van Krevelen, D. W., 1950. Graphical-statistical method for the study of structure and
reaction processes of coal. Fuel 29, 269–284.
Wakeham, S.G., Lee, C., Hedges, J.I., Hernes, P.J., Peterson, M.J., 1997. Molecular indicators
of diagenetic status in marine organic matter. Geochimica et Cosmochimica Acta 61,

of
24, 5363–5369. doi:10.1016/S0016-7037(97)00312-8.
Wang, X. C., Druffel, E.R.M., Griffin, S., Lee, C., Kashgarian, M., 1998. Radiocarbon

ro
Studies of Organic Compound Classes in Plankton and Sediment of the Northeastern
-p
Pacific Ocean. Geochimica et Cosmochimica Acta 62, 8, 1365–1378. doi:10.1016/S0016-
7037(98)00074-X.
re
Wiles, E., Green, A., Watkeys, M., Botes, R., Jokat, W., 2019. Submarine canyons of NW
Madagascar: A first geomorphological insight. Deep Sea Research Part II: Topical Studies
lP

in Oceanography 161, 5-15. doi:10.1016/j.dsr2.2018.06.003.

Yeh, H.W., and Wang, W.M. 2001. Factors affecting the isotopic composition of organic
na

matter. (1) Carbon isotopic composition of terrestrial plant materials. Proceedings of the
National Science Council, Republic of China. Part B, Life sciences, 25(3), 137-147.
ur

Zindorf, M., Rooze, J., Meile, C., März, C., Jouet, G., Newton, R., Brandily, C., and Pastor,
L. 2021. The evolution of early diagenetic processes at the Mozambique margin during the
Jo

last glacial-interglacial transition. Geochimica et Cosmochimica Acta, 300, 79-94.

doi:10.1016/j.gca.2021.02.024

30
 Journal Pre-proof

7. FIGURES

of
ro
-p
re
lP
na
ur
Jo

Fig. 1. Location map of the samples collected in the Mozambique Channel. The geological
context is from Fierens et al. (2019). BDI: Bassas da India Island. EUR: Europa Island. JN:
Juan de Nova. White dots and black letters indicate core sediments (CS, CSF, KSF and KS).
White dots and green letters represent seafloor sediments (MTB). Red stars and red letters
indicate the location of particle traps (MLP).

31
 Journal Pre-proof

of
ro
-p
re

Fig. 2. (A) Cross-plot of Total Carbon determined from Rock-Eval and from LECO elemental
lP

analysis. (B) Comparison of calcite content (%) obtained by Rock-Eval method (Pillot et al.,
2015) and Ca content from XRF analyses for a selected core sample (Pamela-MOZ04-CS17).
na
ur
Jo

32
 Journal Pre-proof

of
ro
-p
Fig. 3. Rock-Eval 6 results: (a) peak S2 vs Total Organic Carbon (TOC) content and (b)
Hydrogen Index (HI) vs Oxygen Index (OI) values, obtained from sedimentary organic matter
re
of suspended sediments collected from the different physiographic provinces of the
Mozambique Channel (Fig. 1).
lP
na
ur
Jo

33
 Journal Pre-proof

of
ro
-p
Fig. 4. Isotopic signature (δ13C) of the organic carbon collected in particle traps (Fig. 1).
re
Values are given as average for each sediment trap (Table 2). Horizontal lines represent the
standard deviation.
lP
na
ur
Jo

34
 Journal Pre-proof

of
ro
Fig. 5. Evolution of OC (%) vs depth from Leco analyzer for seafloor sediments (MTB)
-p
collected in the distal depositional areas.
re
lP
na
ur
Jo

35
 Journal Pre-proof

of
ro
-p
Fig 6. Rock-Eval 6 results (S2 vs TOC) of sedimentary organic matter from cores collected in
the Mozambique channel, in comparison with Orinoco data range (from Deville et al., 2015)
re
and Congo data range (from Baudin et al., 2017). Core locations are given in Figure 1.
lP
na
ur
Jo

36
 Journal Pre-proof

of
ro
Fig 7. TOC (%) from RE6 analyses vs depth (m) in sediment cores sampled in the Zambezi
turbidite system (Zambezi slope: MOZ4-CS17, MOZ4-CSF18, MOZ4-CSF19; Zambezi
-p
valley: MOZ1-KSF23, MOZ4-CSF20; Bassas da India: MOZ4-CS22; Tsiribihina valley:
MOZ1-KSF24, MOZ4-CS24; Distal depositional area: MOZ1-KSF30, MOZ4-CS25, MOZ4-
re
CS26) (for location map see Fig. 1). Note the strong TOC decrease towards the terminal distal
lP

channel.
na
ur
Jo

37
 Journal Pre-proof

of
ro
-p
re
lP

Fig. 8. Total Organic Carbon (TOC) and Inorganic Carbon (MINC) profiles vs depth from core
na

samples in the Majunga margin (for sample locations see Fig. 1).
ur
Jo

38
 Journal Pre-proof

of
ro
-p
re
lP
na
ur
Jo

Fig. 9. Total Organic Carbon (TOC) and Inorganic Carbon (MINC) profiles vs depth along the
Zambezi turbidite system (Mozambique continental slope and deep basin) (for sample
locations see Fig. 1).

39
 Journal Pre-proof

of
ro
-p
re
lP
na

Fig. 10. SEM images illustrating the textural differences in the modern sediments from the
ur

Majunga slope, NW Madagascar (A) and from the Zambezi slope (B) (location is given in
Jo

Fig. 1).

40
 Journal Pre-proof

of
ro
-p
re
lP

Fig. 11. EDS-SEM mineralogical mapping obtained by a statistical calculation (k-means

clustering) on a sample of core Pamela-MOZ04-CS17 (depth 78 cm bsf), Zambezi slope (for
na

sample location see Fig. 1).

ur
Jo

41
 Journal Pre-proof

of
ro
-p
re
Fig. 12. SEM images illustrating (A) the presence of neo-formed iron oxides inside
foraminifera test, (B) the presence of silica in OM, and (C) calcified OM. Samples from core
lP

Pamela-MOZ04-CS17, Zambezi valley (for sample location see Fig.1).

na
ur
Jo

42
 Journal Pre-proof

of
ro
Fig. 13. Mineralogical mapping of a sample of core Pamela-MOZ01-KSF12 (depth 54 cm
bsf), Majunga slope, NW Madagascar (for sample location see Fig. 1).
-p
re
lP
na
ur
Jo

43
 Journal Pre-proof

of
ro
-p
re
lP

Fig. 14. Oxygen profiles measured on interface cores collected in the Mozambique Channel
(for sample locations see Fig. 1).
na
ur
Jo

44
 Journal Pre-proof

of
ro
-p
Fig. 15. Rock-Eval 6 results obtained on sedimentary organic matter from core sediments,
re
seafloor sediments and suspended sediments collected in the Mozambique Channel. a) S2 vs
TOC; b) HI vs OI.
lP
na
ur
Jo

45
 Journal Pre-proof

of
Fig. 16. TOC (%) as a function of water depth (m) along the Zambezi turbidite system. TOC
is presented as average value in each core, with minimum and maximum values (Table 1).

ro
-p
re
lP
na
ur
Jo

46
 Journal Pre-proof

of
ro
-p
re
lP

Fig. 17. Average TOC (%) for representative suspended sediments (orange bar), seafloor
sediments (yellow bar) and core sediments (black bar) in the main physiographic provinces of
the Mozambique Channel: (A) Majunga slope (735 – 789 m water depth); (B) Tsiribihina
na

valley (3089 – 3415 m water depth); (C) Distal depositional area (4054 – 4130 m water
depth). The bathymetric map of the South East Indian Ocean shows the main circulation
ur

patterns observed the Mozambique Channel (modified after Miramontes et al. 2019) (for
Jo

sample locations see Fig. 1).

47
 Journal Pre-proof

of
ro
-p
re
lP
na
ur
Jo

Table 1. Rock Eval results for the core samples collected in the Mozambique channel.

48
 Journal Pre-proof

Table 2. Average OC (%) measured with Leco analyzer and isotopic signature of the
sedimentary organic carbon for specific suspended sediments. MLP = trap location number,
OC = organic carbon, δ13C = isotopic signature with standard deviation, nb = number of

of
measurements.

ro
-p
re
lP
na
ur
Jo

49
 Journal Pre-proof

of
ro
-p
re
lP
na
ur
Jo

Table 3. Qualitative study of the mineralogical composition of samples MOZ01-KSF12-S1-

54 Madagascar NW slope and MOZ04-CS17-S1-78 Zambezi slope (in normalized %).

50
 Journal Pre-proof

of
ro
-p
re
lP
na
ur
Jo

Table 4. OPD: Oxygen penetration depths measured on interface cores; SAR: sediment
210
accumulation rates based mainly on Pbxs profiles in interface cores (NW Madagascar slope
14
and Mozambique slope), on C dating (Zambezi valley) and from an estimation based on
mass accumulation rates in a moored sediment trap (Tsihiribihina valley; see explanation in
the text); OET: oxygen exposure times calculated in interface cores;

a. From Fontanier et al., 2016; b. From Pastor et al., 2020; c. This study (210Pbxs profiles are
plotted in the supplementary Fig. S2, Appendix A).

51
 Journal Pre-proof

of
ro
-p
re
lP
na
ur
Jo

52
 Journal Pre-proof

Declaration of interests

☒ The authors declare that they have no known competing financial interests or personal
relationships that could have appeared to influence the work reported in this paper.

☐The authors declare the following financial interests/personal relationships which may be
considered as potential competing interests:

of
ro
-p
re
lP
na
ur
Jo

53
 Journal Pre-proof

ater Mean Mean Mean Mean

pth S2 Mean S3 TOC TOCmin TOCmax Tmax HI Mean OI nb. Meas.
mg HC/g mg CO2/g mg HC/ mg CO2/g
m rock rock %Wt %Wt %Wt °C g TOC TOC
28 0.88 4.72 1.06 0.8 1.37 410 79.9 441.2 12
60 0.69 4.6 0.98 0.87 1.09 391 70.8 472.8 10
73 1 3.99 1.29 0.83 1.47 409 80.9 348 10
06 2.46 3.56 1.44 0.3 2.08 404 156 272 3
35 0.75 4.73 1.06 0.87 1.62 399 71.6 464 96
618 1.02 5.58 0.79 0.29 1.94 402 120.3 771.7 8
552 0.53 3.94 0.36 0.15 0.77 395 151 709 7
730 0.69 3.39 0.66 0.3 1.2 417 102.2 437 9
206 0.2 3.43 0.21 0.11 0.28 357 12
084 0.22 2.78 0.31 0.13 0.61 381 64 554 9

of
076 0.12 2.28 0.13 0.11 0.19 364 9
57 0.77 4.87 0.98 0.58 0.87 392 78 501 11
98 0.76 4.83 0.78 0.24 1.11 378 122 752 11

ro
909 0.43 3.17 0.2 0.1 0.44 398 26
02 0.5 3.46 1.13 0.49 1.72 -p 344 45 314 28
10 0.76 3.43 0.76 0.7 0.99 379 100 451 18
15 0.28 2.48 0.62 0.5 0.79 369 42 375 10
501 0.31 4.36 0.37 0.19 0.47 367 10
re
036 0.17 1.49 0.34 0.2 0.76 366 46 424 21
099 0.2 2.87 0.31 0.04 0.99 344 42 364 20
lP

089 0.16 2.6 0.25 0.05 0.56 364 42 361 30

388 0.03 1.39 0.06 0.01 0.18 259 10
130 0.05 1.42 0.06 0.01 0.15 339 14
na

81 4.72 5.22 2.25 1.94 3.3 400 207 232 20

415 5.73 6.03 2.49 2.2 2.73 414 229 243 22
ur

054 4.31 4.63 2.1 1.52 2.67 409 224 245 18

833 5.74 8.83 2.43 1.63 3.36 414 233 373 23
450 4.69 8.01 1.88 1.22 2.83 415 248 451 18
Jo

89 1.78 4.03 1.24 1.05 1.43 386 141 311 2

40 1.99 5.27 0.9 409 1
205 0.455 3.56 0.28 0.25 0.31 371 2
082 0.31 3.03 0.21 0.17 0.24 369 2
074 0.25 1.22 0.15 0.1 0.2 345 4
Table 1. Rock Eval results for the core samples collected in the Mozambique channel.

54
 Journal Pre-proof

Table 2. Average OC (%) measured with Leco analyzer and isotopic signature of the
sedimentary organic carbon for specific suspended sediments. MLP = trap location number,
OC = organic carbon, δ13C = isotopic signature with standard deviation, nb = number of
measurements.

nb
MLP OC δ13C Meas.
(%) (‰)
3 2.32 -21.75 ± 0.21 2
5 2.36 -21.93 ± 0.37 3

of
7 1.84 -21.79 ± 0.31 23
8 2.28 -21.69 ± 0.21 22

ro
-p
re
lP
na
ur
Jo

55
 Journal Pre-proof

Table 3. Qualitative study of the mineralogical composition of samples MOZ01-KSF12-S1-

54 Madagascar NW slope and MOZ04-CS17-S1-78 Zambezi slope (in normalized %).

Quartz Calcite Halite Feldspars Kaolinite Dolomite Mica

KSF12-S1-
9.2 28.4 2.1 3.5 22.6 tr 8.7
54
CS17-S1-
19.9 2.1 1.1 16.7 39.2 tr 20.1
78

of
ro
-p
re
lP
na
ur
Jo

56
 Journal Pre-proof

Table 4. OPD: Oxygen penetration depths measured on interface cores; SAR: sediment
210
accumulation rates based mainly on Pbxs profiles in interface cores (NW Madagascar slope
14
and Mozambique slope), on C dating (Zambezi valley) and from an estimation based on
mass accumulation rates in a moored sediment trap (Tsihiribihina valley; see explanation in
the text); OET: oxygen exposure times calculated in interface cores;

a. From Fontanier et al., 2016; b. From Pastor et al., 2020; c. This study ( 210Pbxs profiles are

Area Interface core Lat Long Water depth OPD

(°S) (°E) (m) (mm)
NW Madagascar margin MOZ1-MTB3 15°21.695 45°57.3886 757 34c
MOZ1-MTB8 740 10c

of
15°21.815 45°57.6485
MOZ1-MTB6 15°31.148 45°42.9309 789 17.5a
5 ± 2c

ro
MOZ4-MTB1 15°21.812 45°57.628 735
MOZ4-MTB2 15°21.685 45°57.378 754 28 ± 4c
MOZ4-MTB3 762 5 ± 1c
-p
15°22.230 45°57.110
Mozambique margin MOZ4-MTB4 19°50.650 36°30.765 412 24 ± 4c
MOZ4-MTB5 19°23.330 36°52.390 316 11 ± 2c
re
Zambezi channel MOZ1-MTB10 21°31.835 41°20.7374 3205 83a
Tsiribihina channel MOZ1-MTB12 21°31.105 41°51.6780 3082 110a
lP

Zambezi distal channel MOZ1-MTB15 25°25.612 41°35.7334 4074 120c

plotted in the supplementary Fig. S2, Appendix A).

na
ur
Jo

57
 Journal Pre-proof

HIGHLIGHTS

 Origin and deposition of the Organic Matter in the main physiographic provinces of
the Mozambique Channel
 Rock-Eval 6 data and δ13C isotopic signature to determine the variability of OM
composition within suspended, seafloor and core sediments
 EDS-SEM analysis, Sediment Accumulation Rates and Oxygen Exposure Times to
assess the preservation condition of the OM in the Mozambique Channel
 Important bottom currents are prone to maintain higher OET by the remobilization of
the uppermost sediments and its OM

of
 Small concentration of terrestrial OC (TOC < 0.5%) are preserved in seafloor
sediments of deep-water domain

ro
-p
re
lP
na
ur
Jo

58