Access to this full-text is provided by Wiley.
Content available from Journal of Geophysical Research: Solid Earth
This content is subject to copyright. Terms and conditions apply.
Integrated Geophysical‐Petrological 3D‐Modeling of the
West and Central African Rift System and Its Adjoining
Areas
Estelle Eric Fosso Téguia M
1,2
, Jörg Ebbing
1
, Peter Haas
3
, and Wolfgang Szwillus
1
1
Institute of Geosciences, Kiel University, Kiel, Germany,
2
Institute for Mining and Geological Research, Yaounde,
Cameroon,
3
GEOMAR Helmholtz Centre for Ocean Research, Kiel, Germany
Abstract This study addresses the lithospheric structure of the West and Central African rift system
(WCARS) and explores its origin and development in relation to the enigmatic Cameroon volcanic line (CVL).
Based on a recent seismic tomography model, we subdivide the areas in tectonic domains. We perform
integrated 3D geophysical and petrological forward modeling. By exploring the thickness and composition of
different domains, we compare the model response to the observed topography and gravity anomalies, under
consideration of the available seismic Moho depth points. Our model reveals three distinct domains within the
study area: The WCARS is predominantly underlain by a Phanerozoic‐type lithospheric mantle, surrounded by
the West African and the Congo Cratons, where the lithospheric mantle has a Proterozoic‐type signature.
Between these domains, we identify a transition area where lithospheric thickness changes rapidly. Our
preferred model shows significant variability of crustal thickness from 20 km in the rift area to 50 km beneath
the cratons accompanied by thin lithosphere of 80 km in the rift area to thick lithosphere of up to 240 km beneath
the cratons. The final model confirms that the WCARS' origin is passive, and suggests that the origin of the
CVL, particularly its continental part, is the result of two tectonic events: (a) V‐shaped opening of the
lithospheric mantle beneath the WCARS, resulting in (b) a strong variation of the lithosphere thickness at the
transition between the rift zone and the northwestern part of the Congo craton.
Plain Language Summary In this study, we describe the current structure of the subsurface (from
the surface to a depth of 300 km) in Central and Western Africa. The aim is to understand the formation of the
Central African Rift zone during the opening of the Atlantic Ocean, and how this relates to the linear chain of
volcanoes that cross Cameroon, known as the Cameroon Volcanic Line. To achieve these objectives, we divide
the study area into tectonic domains reflecting their seismological signature, and then, establish a three‐
dimensional representation of the subsurface structure, based on fitting topography and gravity data. Our model
confirms the geological subdivision of the study area into three blocks corresponding to two cratons and a rift
zone, with transitional areas between them. Our model is compatible with a passive origin of the rifts in the
region. We propose that the origin of the volcanic line of Cameroon is related to magma ascent during the
separation of the African and South American plate in connection with the opening of the Atlantic and
channeled by the lithospheric architecture.
1. Introduction
Since the Proterozoic, the African continent has experienced a long and complex geodynamic history with various
geological processes contributing to the current lithospheric architecture. The tectonic activity has been centered
around three major rift systems: the East African Rift System (EARS), the West African Rift System (WARS),
and the Central African Rift System (CARS). Although there is a junction area between the CARS and the EARS
at the Gregory Rift in eastern Africa, the EARS seems not to be connected to the two others: The Cenozoic EARS
is a younger and less mature rift system compared to the Mesozoic CARS and WARS (Globig et al., 2016), and it
is also well known as an active rift system resulting from the ongoing separation of the African and Arabian
plates. For the CARS and WARS, the origin after the break‐up of the Gondwana supercontinent during Jurassic‐
Cretaceous times is still debated (Bosworth, 1992; Fairhead, 2023; Schull, 1988; etc.). These are often jointly
referred to as the West and Central African Rift System (WCARS), a large scale tectonic feature distinctive in the
sense that it traverses the entire continent from west to the east and from the center to the north of the continent. It
adjoins the northern border of the Congo craton, the eastern border of the West African craton, and spans across
RESEARCH ARTICLE
10.1029/2024JB029226
Key Points:
•We present a new 3D model of the
lithosphere for the West and Central
African Rift System (WCARS)
•Our model confirms that the WCARS
has a passive origin
•Our model suggest that the origin of the
Cameroon volcanic line is linked to the
architecture of the WCARS and
adjoining cratons
Correspondence to:
E. E. Fosso Téguia M,
eric.mousse@ifg.uni-kiel.de
Citation:
Fosso Téguia M, E. E., Ebbing, J., Haas, P.,
& Szwillus, W. (2024). Integrated
geophysical‐petrological 3D‐modeling of
the West and Central African Rift System
and its adjoining areas. Journal of
Geophysical Research: Solid Earth,129,
e2024JB029226. https://doi.org/10.1029/
2024JB029226
Received 3 APR 2024
Accepted 13 JUN 2024
Author Contributions:
Conceptualization: Estelle Eric Fosso
Téguia M, Jörg Ebbing
Data curation: Estelle Eric Fosso
Téguia M
Formal analysis: Estelle Eric Fosso
Téguia M
Funding acquisition: Jörg Ebbing
Investigation: Estelle Eric Fosso Téguia
M, Peter Haas, Wolfgang Szwillus
Methodology: Estelle Eric Fosso Téguia
M, Jörg Ebbing
Resources: Estelle Eric Fosso Téguia M,
Jörg Ebbing, Peter Haas,
Wolfgang Szwillus
Software: Estelle Eric Fosso Téguia M
Supervision: Jörg Ebbing
Validation: Estelle Eric Fosso Téguia M,
Jörg Ebbing
Writing – original draft: Estelle
Eric Fosso Téguia M
Writing – review & editing: Estelle
Eric Fosso Téguia M, Jörg Ebbing,
Peter Haas, Wolfgang Szwillus
© 2024 The Authors.
This is an open access article under the
terms of the Creative Commons
Attribution‐NonCommercial License,
which permits use, distribution and
reproduction in any medium, provided the
original work is properly cited and is not
used for commercial purposes.
FOSSO TÉGUIA M ET AL. 1 of 23
parts of the Sahara meta craton (SMC). Consequently, unveiling the lithospheric structure of the WCARS
(Figure 1) is a key element in understanding the geodynamic history of the African continent since the Mesozoic.
How the WCARS relates to the Cameroon volcanic line (CVL) also remains enigmatic, despite multiple studies
dedicated to its origin (Elsheikh et al., 2014; King & Ritsema, 2000; Koch et al., 2012; Milelli et al., 2012; etc.).
Our study examines the origin and evolution of the WCARS and the CVL. For that we investigate the lithospheric
structure of the WCARS by applying integrated geophysical and petrological 3D modeling. Hereby, several
existing regional and local data sets are utilized. As preparation for the modeling, seismic tomography data is used
in a cluster analysis to define tectonic domains in the lithospheric mantle. These tectonic domains are considered
in the 3D forward modeling to define composition and temperature of the lithospheric mantle.
2. Earlier Geophysical Studies
In the past, a number of geophysical and geodynamic studies have been addressing the origin and the evolution of
the WCARS (Bosworth, 1992; Fairhead, 2023; Schull, 1988; etc.). In one of the earliest studies, based on the
interpretation of gravity and seismological data, Fairhead (1986) showed that although both WCARS and EARS
have appeared after extensional tectonics, they exhibit significant tectonic disparities: the EARS experiences
uplift and is characterized by volcanic activity, while the WCARS underwent subsidence and is marked by
sediment deposition.
Most of the subsequent studies agreed that the WCARS is a passive rift system, closely related to the extensional
rift basin model of McKenzie (1978). Taking into consideration the thinning of its crust and its early geological
development, it seems that the WCARS shows a closer association with the breakup of the Gondwana super-
continent rather than the EARS (Fairhead, 1988). In support of that concept, Guiraud et al. (1992) concluded that
Figure 1. Simplified tectonic map of the study area. WAC: West African craton, CC: Congo craton, CB: Congo basin, TB:
Taoudeni basin, SMC: Sahara meta craton, LC: Lake Chad, CASZ: Central African shear zone, CVL: Cameroon volcanic
line, WARS: West African rift system, CARS: Central African rift system. MM′and NN′are the cross sections we will
display from the final model, and L their intersection.
Journal of Geophysical Research: Solid Earth
10.1029/2024JB029226
FOSSO TÉGUIA M ET AL. 2 of 23
the Early Cretaceous to Paleogene development of the WCARS is linked to the build‐up of intraplate extensional
stresses during the break‐up of Gondwana, which caused reactivation of pre‐existing zones of lithospheric
weakness. A recent study by Fairhead (2023) confirms this idea, since there is no active upper mantle process
detected so far beneath the region.
But even though most of these studies describe the WCARS as a passive rift in contrast to the active EARS, the
question of its origin and evolution remains challenging due to the presence of the enigmatic CVL. Somewhat
contradictory to the statement by Fairhead (2023), others studies like Plomerova et al. (1993) based on seis-
mology, Burke (2001) and Asaah et al. (2015) based on geochemistry, suggest that parts of the WCARS have been
affected by active asthenospheric processes, particular in the area of the CVL. This is supported by three ob-
servations: First, 140 Ma ago the CVL was positioned directly above the “711” plume (711 indicates latitude 7°N
and longitude 11.5°E); second, the site moved at a right angle to the continental margin and remained in this
position for almost 125 Ma, and third, a new plate‐wide pattern of shallow‐mantle convection was established
when the African plate came to rest, as suggested as well from seismic anisotropy (Koch et al., 2012).
However, the absence of a discernible chronological pattern of volcanism (De Plaen et al., 2014) would rule out a
simple stationary mantle plume. Furthermore, the CVL exhibited active volcanism occurring over an extended
period of time (66 Ma), but with a relatively low plume volume flux, indeed much smaller than for any of the
accepted plumes (e.g., Hawaii) (Kolínskỳ et al., 2020). Hence, there is still no ultimate agreement on the
mechanisms responsible for the formation of the CVL.
Precise knowledge of the lithospheric structure would help to understand the formation of the WCARS and
deduce the origin of the CVL. Several studies based on different seismic techniques were conducted to describe
the current condition of the lithospheric structure over the WCARS. Fitton (1980) suggested a connection be-
tween the CVL and the Benue trough, while Stuart et al. (1985) presented seismic crustal depth variations between
the CVL and the Benue trough and deduced that the differences over the WCARS could be attributed to sub-
stantial crustal extension. Moreover, an examination of the crustal composition beneath the CVL and adjoining
areas, utilizing models based on shear wave velocity, reveals interesting insights. It appears that the crustal
thickness is comparable (ranging from 35 to 39 km) beneath both the CVL and the Pan African Oubanguides Belt
in the south. However, a notable variation occurs in the northern margin of the Congo Craton, where the crust is
significantly thicker (measuring between 43 and 48 km) and characterized by shear wave velocities greater than
4.0 km s
1
in the lower crust. In contrast, the Garoua rift at the northern border of the CVL and the coastal plain
exhibit a thinner crust, measuring between 26 and 31 km (Tokam et al., 2010). Unfortunately, the region is still
lacking complete seismic data coverage (Szwillus et al., 2019), making these conclusions somewhat incomplete.
Various studies based on potential methods have also been done to describe the structure of the WCARS, notably,
gravity data constrained by available seismic Moho depths have been used to determine crustal thickness (Eyike
& Ebbing, 2015; Ghomsi et al., 2022). However, the studies mentioned above solely focused on the crustal
structure of the WCARS but do not consider the influence of the lithospheric mantle, which plays a crucial role in
affecting observable signals such as the gravity field (Afonso et al., 2008).
On the scale of the African continent, Globig et al. (2016) combined elevation, geoid, and thermal analysis to
address the crustal and the lithospheric structure. They propose a lithospheric thickness model with large vari-
ability with deeper LAB related to cratonic domains (up to 230 km) adjoining the WCARS and shallower LAB
(ca. 90 km) related to Mesozoic WCARS rifting domains in agreement with tomography models, and stating that
the most striking result for the crust was the crustal thinning (28–30 km thickness) imaged along the Mesozoic
West and Central African Rift Systems. But although on a large scale the WCARS appears to be a simple area of
major shear zones terminated by orthogonal extensional basins and surrounded by cratons, it remains complicated
in terms of understanding the multiphase evolution of rifting (Fairhead, 2023).
3. Data
Table 1provides an overview of the data sets, which are described in detail below.
3.1. Elevation and Gravity Data
Figure 2shows the topography and gravity data for the study area. Topography data was obtained from the
ETOPO1 model (Amante & Eakins, 2008).
Journal of Geophysical Research: Solid Earth
10.1029/2024JB029226
FOSSO TÉGUIA M ET AL. 3 of 23
To investigate the lithospheric structure, we analyze both the Bouguer anomaly at a height of 10 km and gravity
gradients at satellite height of 225 km. Both these fields are used as they offer a complimentary sensitivity, with
the Bouguer anomaly being more suitable for identifying sources at crustal level, while satellite gravity gradients
are better suited for lithospheric sources (Bouman et al., 2016).
The Bouguer anomaly (Figure 2d) is calculated from the free air anomaly, which is derived using the spherical
harmonic global earth model XGM2019e (Zingerle et al., 2020), with a maximum degree set to 720 (Figure 2b).
The gravitational effect of topography (Figure 2c) was calculated with the “tesseroids” software package (Uieda
et al., 2016).
Hereby, the calculation areas was extended by 3° to avoid boundary effects.
Figure 2. (a) Topography of the study area after ETOPO1 (Amante & Eakins, 2008), (b) Free‐air anomaly (XGM2019e,
Zingerle et al., 2020), (c) Gravity effect of topography, (d) Bouguer anomaly. See text for more details.
Table 1
Data and Models Used During the Modeling Process
Data Description Reference
Elevation: ETOPO1 Model geometry & gravity reduction Amante and Eakins (2008)
Gravity: Spherical harmonic global earth model XGM2019e Observable to calculate gravity and gravity gradients Zingerle et al. (2020)
Seismic Moho depths Initial Moho depth and local constraints Mooney et al. (2010) and Globig et al. (2016)
Seismic tomography model AF2019 Input for cluster analysis Celli et al. (2020)
Global model Crust1.0 Geometry of sedimentary thickness Laske et al. (2013)
Continental model WINTERC‐G Geometry of initial LAB Fullea et al. (2021)
Heat flow Observable used for discussion only Lucazeau (2019)
Journal of Geophysical Research: Solid Earth
10.1029/2024JB029226
FOSSO TÉGUIA M ET AL. 4 of 23
The same procedure was applied to the satellite gravity gradient components, with adjustment of the window of
spherical harmonic degree, to address particularly deeper structures of our model. The satellite gravity gradient
components were taken up to a maximum degree of 300 from XGM2019e, limiting the data to the longer
wavelengths, which satellite data are sensitive to. The topographic effect for each gradient component was then
calculated and applied to the measured gradient data as illustrated in Figure 3.
For the topographic correction, densities of ρ
topo
=2,670 kg.m
3
and ρ
water
=1,030 kg.m
3
were applied to the
topography and bathymetry, respectively.
3.2. Seismic and Seismological Data
The seismic Moho depths shown in Figure 4a are compiled from a combination of active and passive seismic data
sources, specifically the data sets of Mooney et al. (2010) and Globig et al. (2016). By utilizing the kriging
Figure 3. Satellite gravity gradient at the satellite height of 225 km; (a), (d), and (g) are the xx, yy, and zz components of the
observed field, respectively. (b), (e), and (h) are the xx, yy, and zz components of the topographic effect and (c), (f), and
(i) are the xx, yy, and zz components of the topographic corrected gravity gradient.
Journal of Geophysical Research: Solid Earth
10.1029/2024JB029226
FOSSO TÉGUIA M ET AL. 5 of 23
interpolation technique (Szwillus et al., 2019), we generate a preliminary gridded Moho depth (Figure 4b) and its
uncertainty (Figure 4c) as starting point for the modeling. Additionally, these seismic data points were used to
validate the results of our final 3D modeling study although the distribution of these data points is irregular
throughout the study area, with a concentration along the CVL.
In order to define different lithospheric domains, we use the shear wave tomography model AF2019 (Celli
et al., 2020) (Figure 5). The deviations dVs of each model's point have been determined by subtracting to the mean
value from the values of the corresponding horizontal layer. The model is defined from the surface to the
asthenosphere. Here, we use the model only in the lithospheric extension, up to 300 km depth.
3.3. Additional Data
In addition to the interpolated Moho depths, the global models Crust1.0 (Laske et al., 2013) and WINTERC‐G
(Fullea et al., 2021) serve as an initial reference for the model's geometry. Crust1.0 provides sediment thickness
with a 1‐ degree resolution (Figure 6a). In absence of a more detailed data set, we use this model, which is not
bad at the lithospheric scale, as at that scale the model is not very sensitive to the sedimentary layer.
WINTERC‐G provides the initial lithosphere‐asthenosphere boundary in 2° resolution (Figure 6b).
Figure 6c shows available heat flow data over the study area, taken from Lucazeau (2019). Similar as for seismic
Moho depths, data coverage is sparse. These heat flow data are only used for comparison to our model outputs.
4. Method
To address the structure of the WCARS and its surroundings, we first perform cluster analysis to identify different
tectonic domains. Second, we perform the 3D modeling of the lithosphere, taking into account the lateral vari-
ability of the mantle composition, based on tectonic domains previously determined. Figure 7shows a flowchart
of the modeling steps.
4.1. Cluster Analysis
The K‐means clustering algorithm is used to cluster vertical velocity profiles from regional seismic tomography.
That algorithm follows an iterative process which can be subdivided into three steps: (a) Data are initially par-
titioned into K groups around k random centroids. (b) Centroid positions are recalculated by using the simple
Euclidean distance (given by Equation 1), and new links are made to generate new groups, by minimizing the
mean value (given by Equation 2) of the previous distances between objects and their associated centroid. (c) Step
(b) is repeated until no more transfer of objects between groups (Novianti et al., 2017; Weatherill &
Burton, 2009).
d(x,y) =
x1y1)2+x2y2)2+⋯+xpyp)2
√(1)
Figure 4. (a) Represents the combination of available passive and active seismic Moho depth (Globig et al., 2016; Mooney et al., 2010) in the area. (b) Is the interpolated
Moho depth deduced from a, and c the associated uncertainty following the methodology of Szwillus et al. (2019).
Journal of Geophysical Research: Solid Earth
10.1029/2024JB029226
FOSSO TÉGUIA M ET AL. 6 of 23
where xand y, represent velocity‐depth profiles.
Ckj =xkj +x2kj +⋯+xakj
a,j=1,2, …,p(2)
where C
kj
is centroid of group‐k, variable‐j, and athe number of members in the group k.
Figure 6. (a) Sediment thickness (Laske et al., 2013). (b) Initial LAB from WINTERC‐G (Fullea et al., 2021). (c) Heat flow data points (Lucazeau, 2019).
Figure 5. Deviation of the seismic tomography calculated at different depths. Each slice represents the mean value of Versus for the respective depth. The seismic
tomography is taken from the model AF2019 (Celli et al., 2020).
Journal of Geophysical Research: Solid Earth
10.1029/2024JB029226
FOSSO TÉGUIA M ET AL. 7 of 23
E(m1,m2,…, mk)=∑
N
i=1∑
K
k=1
I(xi∈Ck)‖ximk‖2(3)
where m
k
is the mean of cluster C
k
and I(x) is 1 if statement xis true, 0 otherwise.
Furthermore, assessing the adequacy of a partition is highly significant in the context of cluster analysis in
geophysics. The predominant measure utilized to gauge cluster quality (given a known K) is the total within‐
cluster sum of squares, also denoted as squared error or clustering error (E, given by Equation 3) (Likas
et al., 2003). In our case, this unsupervised machine learning technique groups the velocity‐depth profiles derived
from seismic tomography, delineating lithospheric domains and the transitional regions connecting them, on the
base of the proximity of velocity‐depth profiles to the determined centroids. This approach has already been
successfully applied on a global (Lekic & Romanowicz, 2011; Schaeffer & Lebedev, 2015) and continental (Haas
et al., 2021) scale.
4.2. 3D Modeling
We use a lithospheric forward modeling approach based on Litmod3D (Fullea & Afonso, 2009). The key
advantage of LitMod3D lies in its ability to simultaneously address the thermal and compositional state of the
lithosphere. Within LitMod3D, physical properties in the lithospheric mantle are determined based on pressure,
temperature, and bulk composition. This is achieved by solving equations related to heat transfer, thermody-
namics, rheology, geopotential, and isostasy. Through iterative computations, we can estimate density, tem-
perature, and heat flow by finding the best fit between the modeled and observed gravity and topography.
For a comprehensive understanding of LitMod3D and its theoretical foundation, we refer to Afonso et al. (2008)
and Fullea and Afonso (2009).
The main parameters that have to be set are shown in Table 2. The densities for the crust and sediments initial
values and in situ density are calculated in dependency on the pressure coefficient. Thermal parameters are, in
absence of local data, based on global and previous studies (Globig et al., 2016; Puziewicz et al., 2019). The
density structure of the mantle is based on the chosen composition, in turn based on PerPle_X (J. Connolly, 2009;
J. A. Connolly, 2005).
The geometry of the model and the observables used to validate the model are based on the data sets described in
the following section.
Figure 7. Flowchart of the integrated study carried out in this work.
Journal of Geophysical Research: Solid Earth
10.1029/2024JB029226
FOSSO TÉGUIA M ET AL. 8 of 23
5. Results
5.1. Seismological Regionalization
The regionalization was carried out disregarding the offshore area. Choosing the number of clusters is somewhat
arbitrary. We tested between 2 and 6 clusters and found that 3 clusters effectively represent the known geology of
the region. The velocity were considered between 50 and 300 km depth to prevent any skewed results from near‐
surface heterogeneities.
The cluster findings (Figure 8) indicate that the continental region of the study area can be distinctly divided into
three primary blocks, which roughly correspond to the expected tectonic units.
Within each cluster, the vertical velocity profiles are broadly similar. Cluster 1 profiles exhibit an increase of
seismic velocity from 50 to 150 km, followed by a decrease until reaching approximately 220 km. Subsequently,
there is an increase until around 240 km, after which the velocity remains relatively constant until reaching
300 km.
Cluster 2 consists of profiles that remain relatively constant between 50 km and around 70 km. From there, they
experience a gradual decrease in seismic velocity until approximately 80 km, followed by a slight increase until
reaching around 210 km. Finally, there is another increase as they progress toward 300 km.
Figure 8. Clusters of tectonic domains and corresponding velocity depth profiles. WAC: West African craton, CC: Congo
craton, WCARS: West and central African rift systems. The number 1 to 3 represent groups of clusters. The four velocity‐
depth profiles in highlighted colors represent different centroids around which all the other velocity profiles (in gray) are
clustered. Each profile group describes the corresponding colored zone on the tectonic map. Because of their similarity, the
two lower velocity‐depth profiles have been considered as describing the same tectonic unit.
Table 2
Layers' Thermophysical Properties
Layers Density Kg.m
3
Thermal conductivity W.m
1
.K
1
Heat product‐ion rate W.m
3
Pressure coefficient Grueneisen parameter
Sediments 2,500 2.0 9 ∗10
6
8.5 ∗10
10
/
Crust 2,670 2.5 9 ∗10
6
1.5 ∗10
10
/
Mantle / 3.2 10
–8
/ 1.25
Note. The top mantle's density is directly computed by the code PerPle_X, the pressure coefficient for the mantle and the Grueneisen parameter for sediment and crust
layer are negligible.
Journal of Geophysical Research: Solid Earth
10.1029/2024JB029226
FOSSO TÉGUIA M ET AL. 9 of 23
Lastly, cluster 3 comprises profiles that demonstrate an increase from 50 to 80 km, followed by a gradual decrease
throughout the entire range until reaching 300 km.
Comparing the clustered blocks with the tectonic map of the region, it becomes apparent that cluster 1 corre-
sponds to the Congo craton in the southeast and the West African craton in the northwest. Cluster 3 coincides with
the WCARS, cluster 2 lies between the two other clusters and is identified as a transitional area between craton
and rift zone. It exhibits a consistent velocity at the beginning. A notable dissimilarity between the expected
geological distribution (Figure 1) and the clustering result (Figure 8) is the absence of the southern part of the
Saharan metacraton in the cluster analysis output (northeastern section of the study area).
5.2. Model
5.2.1. Model Setup
To conduct a comprehensive analysis of the WCARS and its surroundings using our model, we considered an area
of 3,800 km ×3,800 km with a depth of 300 km. In order to ensure numerical accuracy, we have employed a
lateral discretization of 50 km and a smaller vertical discretization of 2 km.
To establish the initial model geometry, we have utilized elevation, sediment thickness, interpolated Moho as well
as the LAB presented in the data section.
To account for the diverse lithospheric characteristics present within the region, we have incorporated global
geophysical and petrophysical properties that have been adjusted based on specific domains identified through
cluster analysis. The obtained clusters are implemented as a combination of Proterozoic‐Phanerozoic‐Primitive
Mantle types for cratons—rift—offshore areas respectively.
5.2.2. Lithospheric Mantle Composition
At this stage of our study, we made a number of sensitivity tests to determine the ideal type of the lithospheric
mantle composition in the study area. We first considered the lithospheric mantle to be homogeneous, and tested
the Archean, Proterozoic and Phanerozoic types respectively. The models show that a reasonable fit could be
achieved for the Proterozoic and Phanerozoic types, while for the Archean type no satisfying fit to the observed
topography and gravity anomalies can be achieved, without an unreasonable geometry. For Proterozoic litho-
spheric mantle, the modeled Moho is much shallower than the seismic Moho in rift zones, and for Phanerozoic
lithospheric mantle, it is rather too deep under cratons. A similar observation can be made by comparing the
model to heat flow data. The model with a Proterozoic lithospheric mantle results in reasonable heat flow values
for the cratonic areas, while in the rift area the values are rather low. For the model with the Phanerozoic lith-
ospheric mantle, heat flow values for the rift area appear to be reasonable, but they are too high for the cratons.
Details of this sensitivity test are available in the appendix, Figure A3. Based on these observations and the results
of the clustering exercise, we considered the lithospheric mantle as made of Proterozoic and Phanerozoic do-
mains. More specificly, we introduced Phanerozoic‐type lithospheric mantle under the WCARS surrounded by
Proterozoic lithospheric mantle under the CC and WAC. The transition area between the cratons and the rift area
is modeled as a gradual change leading to an intermediate, mixed composition (see Figure 9a). This model has
shown the best fit between modeled and observed values. In the following, we will concentrate on the discussion
of the best fit model.
Figure 10 shows the fit to the observed data of our preferred model. The modeled elevation closely matches the
measured elevation shown in Figure 2a. In areas of high topography the misfit (Figure 10a1) displays some
noticeable variations between modeled and measured elevation. These short‐wavelength differences are related to
small‐scale structures, which our lithospheric‐scale model is not intended to reproduce. In general, the reasonable
data fit is confirmed by the statistical distribution showing an error close to zero for most of the area, with an RMS
value of 137 m.
The trend observed in the modeled Bouguer anomaly (Figure 10b) closely matches the measured Bouguer
anomaly (Figure 2d). However, the modeled anomaly appears to be smoother, for the same reason as mentioned
above. The differences show RMS value of 23 mGal with a slight shift to positive values.
Figures 10c–10e show the xx, yy, and zz components, respectively, of the modeled gravity gradient across the
study area. The discrepancies observed are minimal, especially for the xx and yy components, which range
Journal of Geophysical Research: Solid Earth
10.1029/2024JB029226
FOSSO TÉGUIA M ET AL. 10 of 23
between 0.3 E and 0.3 E. These components exhibit RMS values below 0.12 E. On the other hand, the misfit for
the zz component (Figure 10e1) displays higher values compared to the other two. Upon closer examination, this
misfit appears to follow the same pattern as the Bouguer anomaly misfit (Figure 10b1),
Moreover, the modeled Moho has been sampled at the available seismic Moho depth points. The comparison of
both of them plotted on top of the modeled Moho (Figure 11a) shows a spatially variable but very low difference,
indicating that the final model is in reasonable agreement with the available data.
5.2.3. Crustal Thickness
Figure 11a shows that the WCARS's crust is thinner compared to adjoining cratonic crust. The thinnest crust is
found in the Benue trough, propagating itself northeastward through northern Cameroon (Yola‐Garoua basin),
Lake Chad, and the middle north of Chad, as well as the Termit basin in Niger. Conversely, the thickest crust is
found in the Tibesti region in the northwest of Chad and the Hoggar region in the Southeast of Algeria. While the
CVL's crust is relatively thick, except in the coastal areas of Cameroon (Mount Cameroon area), a stark contrast
can be observed between the CVL and the Benue trough along the Cameroon central‐west border, as well as the
contact between the Adamawa plateau and the Yola‐Garoua basin.
Figure 9. This figure shows two cross sections through the modeled density, temperature and velocity. (a) represents the vertical cross section (MM′, shown in Figure 1)
crossing the WAC, the CC, the Cameroon volcanic line and the Benue trough. (b) displays the vertical cross section (NN′, shown in Figure 1) along the West and Central
African Rift System, going from the Benue trough toward the known north of the Sahara meta craton (SMC). In opposition to a with large variations of LAB, (b) displays
an almost constant LAB from the Atlantic coast to across the SMC. L is the junction point of the cross sections MM′and NN′. Solid and dashed black lines on top of the
model respectively represent the observed and the modeled Bouguer anomaly. The solid and dashed green lines represent the observed and modeled topography. The
dashed red line represent the modeled heat flow.
Journal of Geophysical Research: Solid Earth
10.1029/2024JB029226
FOSSO TÉGUIA M ET AL. 11 of 23
Figure 10. Modeled Topography and gravity and their misfit in comparison to measured data shown in Figures 2and 3.
(a) Topography, (a1) topography misfit, (a2) residual topography diagram, (b) Bouguer anomaly, (b1) Bouguer anomaly
misfit, (b2) residual Bouguer anomaly diagram, (c) gravity gradient xx component, (c1) xx misfit, (c2) residual xx diagram,
(d) gravity gradient yy component, (d1) yy misfit, (d2) residual yy diagram, (e) gravity gradient zz component, (e1) zz misfit,
(e2) residual zz diagram.
Journal of Geophysical Research: Solid Earth
10.1029/2024JB029226
FOSSO TÉGUIA M ET AL. 12 of 23
5.2.4. Lithospheric Thickness
The modeled lithospheric thickness (Figure 11b) shows the expected difference between the cratons and the rift
zone. Within the cratonic zone, the lithospheric thickness ranges from approximately 160–240 km, and around
80–140 km over the rifts. In the transition zones lithospheric thickness deepens from 140 to 160 km. From the
Benue trough to the Marrah region inside Sudan, the lithosphere is thin ‐ as low as 80 km in certain regions like
Marrah mountains. In the NE direction, there is no clear boundary between the rift zone and the SMC. On the east
and west sides of the rift, we can clearly see the transitions that differ depending on whether we go toward the CC
or the WAC. Unlike the transition toward the CC, which is horizontal over a short distance but with a sharp
variation in depth, the transition toward the WAC takes place over a relatively long distance and evolves in the
form of a staircase around a depth of 160 km. This results in an asymmetrical distribution of the two cratons in
relation to the rift.
5.2.5. Heat Flow
The modeled heat flow (Figure 12) follows the lithospheric thickness to a
large degree and differs as well between the colder cratons ranging from
(46 mW · m
2
to 54 mW · m
2
) and the warmer rift zone with values ranging
from 50 mW · m
2
to 64 mW · m
2
. Additionally, the model indicates that the
CVL, with an approximate value of 60 mW · m
2
, experiences higher heat
flow compared to the Benue depression, which has an approximate value of
54 mW · m
2
. Furthermore, the sudden thinning of the lithosphere between
the Adamawa plateau and the Garoua rift is causing a strong increase in heat
flow in the direction of crustal thinning.
5.2.6. Density
Within the study area, there is a noticeable difference in the density distri-
bution between the WCARS and the adjoining cratons, as well as internal
heterogeneities inside the WCARS (Figure 13). Upon examining a horizontal
section of the model at a depth of 50 km, the cratons exhibit a fairly constant
density of around 3,350 kg.m
3
, whereas the rift zone have highly variable
density. The most prominent contrast of approximately 30 kg.m
3
is
observed in the Marrah Mountains region, which spans from western Sudan to
crossing the border with Chad. This deficit in mass appears to extend in a
northeast to southwest direction toward the Gulf of Guinea, passing through
southern Chad where it is less pronounced, before eventually reaching the
Figure 11. (a) Modeled Moho depth and misfit (colored points) in comparison to available seismic Moho depth; (b) Modeled
LAB. The “V” shape of the opening of the lithosphere is highlighted by the dotted lines.
Figure 12. Modeled heat flow and misfit (colored dots) in comparison to
available heat flow points in the region.
Journal of Geophysical Research: Solid Earth
10.1029/2024JB029226
FOSSO TÉGUIA M ET AL. 13 of 23
Adamawa Plateau in central‐western Cameroon and the Jos Plateau in northeastern Nigeria, where the contrast is
once again significant (around 20 kg.m
3
).
Similar low densities are also identified in the Tibesti mountain region, albeit to a lesser extent in the Aïr (Niger)
and Hoggar mountain regions. These areas with lower density correspond to regions with higher topography,
which aligns with isostatic compensation. However, there are certain exceptions such as the Benue trough,
Murzuq Basins, where a minor contrast of approximately 5 kg.m
3
is observed.
At larger depths, the density distribution gradually becomes more uniform for both the WCARS and the cratons.
However, at 150 km depth, a notable heterogeneity emerges in the south‐western region of the WAC, segregating
the south‐western portion of this craton. That peculiar contrast in density is probably related to the variation of
LAB depth. The gap of densities at the edge of the cratons, noticeable after 150 km are probably due to the
compositional boundary (depleted =lighter) and the fact that the temperature near the LAB is similar to
asthenospheric temperatures.
Figure 14 provides a 3D representation of the modeled density distribution, extending from the topographic
surface down to a depth of 300 km. The vertical sections cross the CC, WAC, and WCARS regions. The model
demonstrates a downward‐oriented density gradient, except for the area below the WCARS. Below the WCARS,
the density decreases from the middle of the lithospheric mantle toward the LAB, before increasing again inside
the asthenosphere. That depletion is probably the consequence of the conflict between the high temperature due to
the shallow asthenosphere and the relative low pressure due to the thin lithosphere in that area.
The 2D cross‐section of the density distribution along MM' (passing trough the middle of the previous mentioned
heterogeneity located in the southwest of the WAC, Figure 9a), emphasizes the transitional zone between the
Figure 13. Horizontal 2D sections of the modeled density distribution at 50 km, 100 km, 150 km, and 200 km depth.
Journal of Geophysical Research: Solid Earth
10.1029/2024JB029226
FOSSO TÉGUIA M ET AL. 14 of 23
depleted Proterozoic lithospheric mantle beneath the WAC and the fertile Phanerozoic lithospheric mantle
beneath the WCARS. It emphasizes as well the transitional zone between the fertile Phanerozoic lithospheric
mantle beneath the WCARS and the Proterozoic lithospheric mantle beneath the CC. These transitional areas
exhibit a significant reduction in mass compared to the adjoining blocks. On the other hand, the 2D cross‐section
along the NN' well shows that there is no clear boundary between WCARS and SMC in terms of LAB depth
variation, but also density, temperature and velocity distributions (Figure 9b).
6. Discussion
6.1. Structure of the Lithosphere
The thickness of the crust beneath the WCARS and its surroundings is highly variable. Isostatic compensation can
explain most of these variations, especially in mountainous regions like Marrah (38 km), Tibesti (50 km), Hoggar
(54 km), and the Adamawa plateau (40 km, similar to the findings of Ghomsi et al., 2022). A stark lateral crustal
thickness gradient is located at the boundary between the oceanic domain and WCARS, reflecting the transition
from oceanic to continental crust.
Despite their proximity, the CVL and the Benue trough exhibit significant differences in terms of Moho depth, as
mentioned by Stuart et al. (1985). Our model suggests that the Benue trough has a Moho depth of approximately
27–30 km, while the CVL ranges from 37 to 40 km, consistent with the results of the shear wave velocity study
conducted by Tokam et al. (2010). A similar variation in crustal thickness is observed on the western side of the
Benue trough, extending from the Atlantic Ocean boundary to the Jos Plateau (approximately 33–38 km). This
indicates that the region likely experienced significant tectonic events, with the Benue trough maintaining the
position of symmetrical axis during the rearrangement. Additionally, the crustal thinning observed in the Benue
depression extends northeastward into the interior of the continent underneath sedimentary basins such as Yola,
Garoua, and Chad, which occupy a substantial surface area and align with the findings of Globig et al. (2016), who
also identified crustal thinning (ca. 28–30 km) along the Mesozoic systems of the WCARS. In contrast, the depth
of the Moho is slightly greater under the CC and WAC (ca. 33–43 km).
The lithospheric thickness in the WCARS and its adjoining areas has a large range of variation, from approxi-
mately 80–240 km. This aligns with previous findings (such as Globig et al., 2016) that the lithosphere‐
asthenosphere boundary exhibits significant spatial variability (ranging from 90 to 230 km) within the region.
Figure 14. Perspective view of the 3D density distribution, with geological block and layer boundaries. The depth has been
exaggerated with factor of 7,000.
Journal of Geophysical Research: Solid Earth
10.1029/2024JB029226
FOSSO TÉGUIA M ET AL. 15 of 23
The highest values are associated with the cratonic domains adjoining the WCARS, while the smallest values are
related to the rift domains themselves during the Mesozoic period (Globig et al., 2016).
Furthermore, our model characterizes the lithospheric mantle beneath the WCARS and its surroundings as
predominantly Phanerozoic and Proterozoic‐type. It is important to note that the composition of the sub‐
continental lithospheric mantle is generally not homogeneous across a large area. With regard to WCARS,
this is in agreement with other studies in the region (Bosworth, 1992; Eyike & Ebbing, 2015; Schull, 1988), which
have described the majority of the inner portions of WCARS as Cenozoic and Mesozoic (Phanerozoic) in origin.
Corroborating these descriptions, Fairhead (1988) states that the opening of the Atlantic basins can be divided into
four stages of development. These stages include the Jurassic opening of the Central Atlantic, the Early Creta-
ceous (approximately 130–119 Ma) opening of the South Atlantic with rifts spreading deeply into Africa via the
Benue trough, the opening of the Equatorial Atlantic (119–105 Ma), the connection of these ocean basins, and the
formation of the current mid‐ocean rift system. Additionally, the WCARS underwent development until the end
of the Cretaceous period. While the precise boundaries between WCARS and the adjoining units are not clearly
defined, it is well known that WCARS is adjacent to cratons. Therefore, the classification of the blocks neigh-
boring WCARS as Proterozoic, as described in this study, aligns with Artemieva (2011), who concluded that
Proterozoic cratons are surrounded by Phanerozoic belts.
We find no discernible difference between the lithospheric mantles under the WCARS and SMC (except in the far
North–West part of the SMC). In fact, the cluster analysis of seismic velocity in the study area clearly shows the
distinction between WCARS, CC, and WAC. However, the SMC, as depicted in the geological map (Figure 1),
remains inconspicuous. This implies that the SMC lacks any distinct characteristics of anomalous seismic ve-
locities within the depth range of 50–300 km. Additionally, our modeled LAB indicates no significant variation
between the thin WCARS lithosphere and the SMC lithosphere (Figures 9b and 11b). The density, temperature,
and velocity distributions determined from our modeling efforts align with this observation, which concurs with
the findings of Sobh et al. (2020), who also inferred a thin lithosphere beneath the SMC based on lithospheric
modeling. Multiple geological events, including the partial delamination of the lithospheric mantle during the
Neoproterozoic era, rifting during the Mesozoic‐Cenozoic period, and the activation of the SMC during the
Mesozoic (Sobh et al., 2020) likely facilitated the fertilization of the region through erosion of isolated litho-
spheric roots.
6.2. Origin of the WCARS
Our model suggests an asymmetry in the LAB geometry between the CC and WAC centers with respect to
WCARS. The LAB under the WAC displays a step‐like pattern toward its center, which can indicate a stretching
of the lithosphere in that region. Consequently, during the breakup of Gondwana, the western side of the current
African continent experienced significant traction. However the traction was distributed differently within the
continent, with the western block absorbing most of it. This explains the cascading form of the lithosphere
observed under the eastern margin of the WAC lithosphere in contrast to the transition in one step between the rift
and the northwest margin of the CC. Additionally, the “V” shape of the WCARS also extending across the SMC
despite its pre‐existing suggests that the pulling force that gave rise to the WCARS was most likely greater to the
north, or simply multi‐directional.
The findings of this work further support the characterization of WCARS as a passive rift system (Fair-
head, 2023), consistent with the extensional rift basin model proposed by McKenzie (1978).
Moreover, along the Benue trough to the Yola‐Garoua basin, the stretching factor (see appendix B, stretching
factor's section) is around 1.2 (which indicates a continental rift, Allen & Allen, 2013), while to the northeast
(South SMC) it is reduced to 1.1 (closer to an intracratonic sag basin, Allen & Allen, 2013). As expected, the
cratonic areas have a stretch factor lower than 1 due to their stability in opposition to the Gulf of Guinea, with a
stretch factor higher than 1.5.
In the case of an active origin of the WCARS, we expected to find the area of thinnest lithospheric mantle under
the volcanic line, however the model shows that the LAB beneath the Benue trough is shallower than beneath
the CVL.
Journal of Geophysical Research: Solid Earth
10.1029/2024JB029226
FOSSO TÉGUIA M ET AL. 16 of 23
6.3. Origin of the CVL
Previous research has already noted the connection between CVL and the
Benue trough (Fitton, 1980; Stuart et al., 1985). Our model indicates that the
CVL could be connected with the CC, especially with the shape of the
northwestern boundary of the CC where there is the most pronounced vari-
ation of the LAB (Figure 15) between the rifts and the cratons.
According to our model, the origin of the CVL could be attributed to the
opening of the WCARS in addition to the particular shape of the northwest
margin of the CC, in agreement with the theory of edge convection mentioned
by Milelli et al. (2012); Koch et al. (2012) to describe the cause of the CVL.
We suggest that during the break‐up of the Gondwana supercontinent in the
Jurassic‐Cretaceous period, mantle material from the interior of the continent
(current WCARS) started to flow toward the present‐day Atlantic Ocean that
is, SW movement (Elsheikh et al., 2014), pulled in by the movement of the
South American tectonic plate moving away from the African plate. The “V”
shape of the lithospheric mantle opening caused the flow velocity to increase
from northeast to southwest in relation to the Venturi effect phenomenon
(Şcheaua, 2016). In addition, the northwestern margin of the CC has acted as a
deep frontal barrier, resulting in convective movement, driven not by tem-
perature gradients but by magma dynamics. This mechanism agrees with the
study of King & Ritsema, 2000, who based on numerical modeling to study
the origin of intraplate volcanism in South America and Africa and concluded
that the CVL is the consequence of edge‐driving convection beneath the CC.
Such a temporary event could have led to the formation of magma chambers
along the known CVL. The pressure within each chamber varied depending
on factors such as the amount of water present and the quantity and quality of
minerals being melted, causing the equilibrium between magmatic pressure
and lithostatic pressure to be effectively random (well detailed in Gud-
mundsson, 2012). This randomness could explain the absence of any age
progression along the CVL (De Plaen et al., 2014).
7. Conclusions
This study examined the lithospheric structure of the WCARS and its sur-
roundings and the link to the enigmatic CVL, by applying integrated
geophysical and petrological 3D modeling.
Cluster analysis of shear wave tomography grouped the study area into three main tectonic blocks. Namely the
rift zone, the northern part of the Congo craton and the east part of the West African craton. Furthermore, the
clustering highlights the absence of any particular signature from the southern part of the Saharan meta craton.
Initial models with a uniform mantle composition were not able to explain all observables simultaneously
(topography, gravity, Moho depth, heat flow). Therefore, in our preferred model, a Phanerozoic‐type litho-
spheric mantle under the WCARS is linked to the two adjoining cratons of Proterozoic‐type by transitional
areas of different size and shape. Further findings of this study confirm the wide range of variations in crustal
(ca. 25–50 km) and lithospheric (ca. 70–250 km) thicknesses across the study area, with the shallowest LAB
found under the region of the Benue trough‐Cameroon volcanic line until the Marrah region, and the deepest
under cratonic centers. They also confirms the absence of any cratonic characteristics of the Sahara meta‐craton
(particularly its southern part), suggesting that it has undergone long term fertilization facilitated by tectonic
events.
In addition, our model confirms the suggestion that the WCARS is a passive rift system. This observation and the
overall lithospheric structure suggests that the origin of the CVL, specially its continental part, could be explained
taking into account the V‐shaped opening of the lithospheric mantle beneath the West‐Central African rift system,
along with the abrupt transition between the rift and the northwestern edge of the Congo craton. This potential
Figure 15. Sketch showing magma movement between cratonic roots during
the Atlantic's opening and the West and Central African Rift System
formation. The magma moving from the continent to the passive margin
through the continental rift, and facing the northwestern border of the CC,
leading to magma uploading inside the CVL's lithosphere.
Journal of Geophysical Research: Solid Earth
10.1029/2024JB029226
FOSSO TÉGUIA M ET AL. 17 of 23
development should be explored further in the future with geodynamic modeling, where also the interaction
between mantle flow and the lithospheric structure is studied in more detail.
Appendix A: Homogeneous Lithospheric Mantle Across the Study Area
This section presents the limits of the integrated 3D modeling across the study area, considering the lithospheric
mantle as homogeneous, which means that the cluster analysis hasn't been processed and no particular blocs have
been defined. The investigation focused on three types of mantle namely Archean, Proterozoic and Phanerozoic,
described in Table A1.
A1. Bouguer Anomaly and Topography
Figures A1 and A2 depict the best Bouguer anomaly and topography models for three different homogeneous
mantle compositions, along with their respective misfits compared to the measured values (Figure 2d).
By inspecting the Bouguer anomaly (Figure A1), it is evident that the datafit of the three compositions are more
similar than different however, some variations can be observed in their misfit maps. Particularly, the Archean
compositions exhibit a large and positive anomaly toward the south–east of the study area (Congo basin). Apart
from this discrepancy, the gravity anomaly does not display significant variations between the three mantle types.
Examining the topography, it becomes more evident that the Archean type (Figure A2, A) is not a suitable fit
compared to the other two mantle types. This is evident as its modeled topography strongly underestimates the
known topography, except for mountainous regions. Certain areas such as cratons and parts within the rifts are
situated well below sea level, reaching depths of up to 600 m in some locations, which is unrealistic. The
corresponding misfits and statistical analysis reveal the significant disagreement of the Archean type with the
observed topography, while also highlighting how well the Proterozoic (Figure A2, B) and Phanerozoic
(Figure A2, C) types align with the observed signals.
From a technical standpoint, the gravity and topography results indicate that the Archean lithospheric mantle type
should not be considered for further analysis.
A2. Heat Flow and Moho Depth
Upon further analysis of parameters such as heat flow and Moho depth, significant differences were observed
between the Proterozoic and Phanerozoic lithospheric mantle and the expected values.
Regarding the Proterozoic lithospheric mantle, the heat flow (see Figure A3, A) ranging from 48 mW.m
2
to
52 mW.m
2
in the cratonic zones was deemed reasonable. However, the heat flow across the rift area was very
low and did not align with expectations, as displaying similar values across the Congo craton, except over the
highland zones. On the other hand, the Phanerozoic (Figure A3, B) showcased a suitable range of heat flow
(54 mW.m
2
–64 mW.m
2
) in the rift zone but unsatisfactory values across the cratons, which seemed generally
Table A1
Mantle Composition, Griffin et al. (2009), Fullea and Afonso (2009), and McDonough and Sun (1995)
Wt percentage Archean LM Proterozoic LM Phanerozoic LM
SiO
2
45.7 44.7 44.5
Al
2
O
3
0.99 2.1 3.5
FeO 6.4 7.9 8.0
MgO 45.5 42.4 39.8
CaO 0.59 1.9 3.1
Na
2
O 0.07 0.15 0.24
Note. LM for lithospheric mantle.
Journal of Geophysical Research: Solid Earth
10.1029/2024JB029226
FOSSO TÉGUIA M ET AL. 18 of 23
warmer than the rift zone. As a result, this analysis indicated that the Proterozoic type was more suitable for the
craton areas while the Phanerozoic type was better suited for the rift zone. Additionally, the Moho in too shallow
range across the rifts for the Proterozoic lithospheric mantle and in too deep range across the cratons for the
Phanerozoic lithospheric mantle, compared to the expected values, further reinforced the idea of dividing the
study area based on the type of lithospheric mantle.
In summary, this case study has led to a reevaluation of the study area, no longer considering it as homogeneous
but rather as heterogeneous, with Proterozoic type prevalent across the cratons and Phanerozoic type dominant
across the rift zones.
Figure A1. Modeled Bouguer anomaly for homogeneous mantle, their misfit to the measured Bouguer anomaly (shown in Figure 2) and the associated residual diagram.
(a) represents the result for the Archean type, (b) Proterozoic type, and (c) the Phanerozoic type.
Journal of Geophysical Research: Solid Earth
10.1029/2024JB029226
FOSSO TÉGUIA M ET AL. 19 of 23
Figure A2. Modeled topography for homogeneous mantle, their misfit to the measured topography (shown in Figure 2) and the associated residual diagram.
(a) represents the result for the Archean type, (b) Proterozoic type, and (c) the Phanerozoic type.
Journal of Geophysical Research: Solid Earth
10.1029/2024JB029226
FOSSO TÉGUIA M ET AL. 20 of 23
Appendix B: Stretching Factor
As shown by Svartman Dias et al. (2016); Allen and Allen (2013), the stretching factor s, s=30∗Moho
1
can help
to investigate the deformations that faced the crust. We show below the stretching factor of the crust over the
study area. We can see that range of the stretching factor inside the WCARS is well in agreement with a passive
origin (Figure B1).
Figure B1. Stretching factor.
Figure A3. (a) and (b) are the modeled heat flow in the case of Proterozoic and Phanerozoic types respectively, (c) and
(d) represent the modeled Moho depth in the case of Proterozoic and Phanerozoic types respectively.
Journal of Geophysical Research: Solid Earth
10.1029/2024JB029226
FOSSO TÉGUIA M ET AL. 21 of 23
Data Availability Statement
The data and code files used in this paper are available on Zenodo (Fosso Teguia M et al., 2024a,2024b).
References
Afonso, J. C., Fernandez, M., Ranalli, G., Griffin, W., & Connolly, J. (2008). Integrated geophysical‐petrological modeling of the lithosphere and
sublithospheric upper mantle: Methodology and applications. Geochemistry, Geophysics, Geosystems,9(5). https://doi.org/10.1029/
2007gc001834
Allen, P. A., & Allen, J. R. (2013). Basin analysis: Principles and application to petroleum play assessment. John Wiley & Sons.
Amante, C., & Eakins, B. (2008). Etopo1 1 arc‐minute global relief model: Procedures, data sources and analysis. National Geophysical Data
Center, NESDIS, NOAA, US Dept. Commerce.
Artemieva, I. (2011). Lithosphere: An interdisciplinary approach. Cambridge University Press.
Asaah, A. N., Yokoyama, T., Aka, F. T., Usui, T., Wirmvem, M. J., Tchamabe, B. C., et al. (2015). A comparative review of petrogenetic
processes beneath the Cameroon volcanic line: Geochemical constraints. Geoscience Frontiers,6(4), 557–570. https://doi.org/10.1016/j.gsf.
2014.04.012
Bosworth, W. (1992). Mesozoic and early tertiary rift tectonics in east Africa. Tectonophysics,209(1–4), 115–137. https://doi.org/10.1016/0040‐
1951(92)90014‐w
Bouman, J., Ebbing, J., Fuchs, M., Sebera, J., Lieb, V., Szwillus, W., et al. (2016). Satellite gravity gradient grids for geophysics. Scientific
Reports,6(1), 1–11. https://doi.org/10.1038/srep21050
Burke, K. (2001). Origin of the Cameroon line of volcano‐capped swells. The Journal of Geology,109(3), 349–362. https://doi.org/10.1086/
319977
Celli, N. L., Lebedev, S., Schaeffer, A. J., & Gaina, C. (2020). African Cratonic lithosphere carved by mantle plumes. Nature Communications,
11(1), 1–10. https://doi.org/10.1038/s41467‐019‐13871‐2
Connolly, J. (2009). The geodynamic equation of state: What and how. Geochemistry, Geophysics, Geosystems,10(10). https://doi.org/10.1029/
2009gc002540
Connolly, J. A. (2005). Computation of phase equilibria by linear programming: A tool for geodynamic modeling and its application to subduction
zone decarbonation. Earth and Planetary Science Letters,236(1–2), 524–541. https://doi.org/10.1016/j.epsl.2005.04.033
De Plaen, R., Bastow, I., Chambers, E., Keir, D., Gallacher, R., & Keane, J. (2014). The development of magmatism along the Cameroon volcanic
line: Evidence from seismicity and seismic anisotropy. Journal of Geophysical Research: Solid Earth,119(5), 4233–4252. https://doi.org/10.
1002/2013jb010583
Elsheikh, A. A., Gao, S. S., & Liu, K. H. (2014). Formation of the Cameroon volcanic line by lithospheric basal erosion: Insight from mantle
seismic anisotropy. Journal of African Earth Sciences,100, 96–108. https://doi.org/10.1016/j.jafrearsci.2014.06.011
Eyike, A., & Ebbing, J. (2015). Lithospheric structure of the west and central African rift system from regional three‐dimensional gravity
modelling. South African Journal of Geology,118(3), 285–298. https://doi.org/10.2113/gssajg.118.3.285
Fairhead, J. (1986). Geophysical controls on sedimentation within the African rift systems (Vol. 25, pp. 19–27). Geological Society, London,
Special Publications. https://doi.org/10.1144/gsl.sp.1986.025.01.03
Fairhead, J. (2023). The mesozoic west and central africa rift system (WCARS) and the older kandi shear zone (KSZ): Rifting and tectonics of
north africa and south America and fragmentation of gondwana based on geophysical investigations. Journal of African Earth Sciences,199,
104817. https://doi.org/10.1016/j.jafrearsci.2022.104817
Fairhead, J. D. (1988). Mesozoic plate tectonic reconstructions of the central South Atlantic Ocean: The role of the West and central African rift
system. Tectonophysics,155(1–4), 181–191. https://doi.org/10.1016/0040‐1951(88)90265‐x
Fitton, J. (1980). The Benue trough and Cameroon line—A migrating rift system in west Africa. Earth and Planetary Science Letters,51(1), 132–
138. https://doi.org/10.1016/0012‐821x(80)90261‐7
Fosso Teguia M, E. E., Ebbing, J., Haas, P., & Szwillus, W. (2024a). Integrated geophysical‐petrological 3d‐modelling of the west and central
African rift system and its adjoining areas (version 2) [Dataset]. Zenodo.https://doi.org/10.5281/zenodo.8399175
Fosso Teguia M, E. E., Ebbing, J., Haas, P., & Szwillus, W. (2024b). Integrated geophysical‐petrological 3d‐modelling of the west and central
African rift system and its adjoining areas (version 1) [Software]. Zenodo.https://doi.org/10.5281/zenodo.11207685
Fullea, J., & Afonso, J. C. (2009). Litmod3d users guide.
Fullea, J., Lebedev, S., Martinec, Z., & Celli, N. (2021). Winterc‐g: Mapping the upper mantle thermochemical heterogeneity from coupled
geophysical–petrological inversion of seismic waveforms, heat flow, surface elevation and gravity satellite data. Geophysical Journal In-
ternational,226(1), 146–191. https://doi.org/10.1093/gji/ggab094
Ghomsi, F. E. K., Tenzer, R., Njinju, E., & Steffen, R. (2022). The crustal configuration of the west and central African rift system from gravity
and seismic data analysis. Geophysical Journal International,230(2), 995–1012. https://doi.org/10.1093/gji/ggac089
Globig, J., Fernàndez, M., Torne, M., Vergés, J., Robert, A., & Faccenna, C. (2016). New insights into the crust and lithospheric mantle structure
of Africa from elevation, geoid, and thermal analysis. Journal of Geophysical Research: Solid Earth,121(7), 5389–5424. https://doi.org/10.
1002/2016jb012972
Griffin, W., O’reilly, S. Y., Afonso, J. C., & Begg, G. (2009). The composition and evolution of lithospheric mantle: A re‐evaluation and its
tectonic implications. Journal of Petrology,50(7), 1185–1204. https://doi.org/10.1093/petrology/egn033
Gudmundsson, A. (2012). Magma chambers: Formation, local stresses, excess pressures, and compartments. Journal of Volcanology and
Geothermal Research,237, 19–41. https://doi.org/10.1016/j.jvolgeores.2012.05.015
Guiraud, R., Binks, R., Fairhead, J., & Wilson, M. (1992). Chronology and geodynamic setting of cretaceous‐cenozoic rifting in west and central
Africa. Tectonophysics,213(1–2), 227–234. https://doi.org/10.1016/0040‐1951(92)90260‐d
Haas, P., Ebbing, J., Celli, N. L., & Rey, P. F. (2021). Two‐step gravity inversion reveals variable architecture of African cratons. Frontiers in
Earth Science,1240.https://doi.org/10.3389/feart.2021.696674
King, S. D., & Ritsema, J. (2000). African hot spot volcanism: Small‐scale convection in the upper mantle beneath cratons. Science,290(5494),
1137–1140. https://doi.org/10.1126/science.290.5494.1137
Koch, F. W., Wiens, D. A., Nyblade, A. A., Shore, P. J., Tibi, R., Ateba, B., et al. (2012). Upper‐mantle anisotropy beneath the Cameroon volcanic
line and Congo craton from shear wave splitting measurements. Geophysical Journal International,190(1), 75–86. https://doi.org/10.1111/j.
1365‐246x.2012.05497.x
Acknowledgments
This study has been funded by the
Deutsche Forschungsgemeinschaft (DFG,
German Research Foundation)—
441292957. We thank the editors and the
reviewers for their constructive comments
on the earlier version which has permitted
to greatly improve the manuscript. Open
Access funding enabled and organized by
Projekt DEAL.
Journal of Geophysical Research: Solid Earth
10.1029/2024JB029226
FOSSO TÉGUIA M ET AL. 22 of 23
Kolínskỳ, P., Schneider, F. M., & Bokelmann, G. (2020). Surface wave diffraction pattern recorded on Alparray: Cameroon volcanic line case
study. Journal of Geophysical Research: Solid Earth,125(7), e2019JB019102. https://doi.org/10.1029/2019jb019102
Laske, G., Masters, G., Ma, Z., & Pasyanos, M. (2013). Update on crust1. 0—A 1‐degree global model of earth’s crust. Geophysical Research
Abstracts,15, 2658.
Lekic, V., & Romanowicz, B. (2011). Tectonic regionalization without a priori information: A cluster analysis of upper mantle tomography. Earth
and Planetary Science Letters,308(1–2), 151–160. https://doi.org/10.1016/j.epsl.2011.05.050
Likas, A., Vlassis, N., & Verbeek, J. J. (2003). The global k‐means clustering algorithm. Pattern Recognition,36(2), 451–461. https://doi.org/10.
1016/s0031‐3203(02)00060‐2
Lucazeau, F. (2019). Analysis and mapping of an updated terrestrial heat flow data set. Geochemistry, Geophysics, Geosystems,20(8), 4001–
4024. https://doi.org/10.1029/2019gc008389
McDonough, W. F., & Sun, S.‐S. (1995). The composition of the earth. Chemical Geology,120(3–4), 223–253. https://doi.org/10.1016/0009‐
2541(94)00140‐4
McKenzie, D. (1978). Some remarks on the development of sedimentary basins. Earth and Planetary Science Letters,40(1), 25–32. https://doi.
org/10.1016/0012‐821x(78)90071‐7
Milelli, L., Fourel, L., & Jaupart, C. (2012). A lithospheric instability origin for the Cameroon volcanic line. Earth and Planetary Science Letters,
335, 80–87. https://doi.org/10.1016/j.epsl.2012.04.028
Mooney, W., Romanowicz, B., & Dziewonski, A. (2010). 1.11 crust and lithospheric structure–global crustal structure. Seismology and Structure
of the Earth: Treatise on Geophysics,1, 361.
Novianti, P., Setyorini, D., & Rafflesia, U. (2017). K‐means cluster analysis in earthquake epicenter clustering. International Journal of Advances
in Intelligent Informatics,3(2), 81–89. https://doi.org/10.26555/ijain.v3i2.100
Plomerova, J., Babuška, V., Dorbath, C., Dorbath, L., & Lillie, R. (1993). Deep lithospheric structure across the central African shear zone in
Cameroon. Geophysical Journal International,115(2), 381–390. https://doi.org/10.1111/j.1365‐246x.1993.tb01193.x
Puziewicz, J., Czechowski, L., Grad, M., Majorowicz, J., Pietranik, A., & Šafanda, J. (2019). Crustal lithology vs. thermal state and Moho heat
flow across the NE part of the European Variscan orogen: A case study from SW Poland. International Journal of Earth Sciences,108(2), 673–
692. https://doi.org/10.1007/s00531‐018‐01674‐7
Schaeffer, A., & Lebedev, S. (2015). Global heterogeneity of the lithosphere and underlying mantle: A seismological appraisal based on
multimode surface‐wave dispersion analysis, shear‐velocity tomography, and tectonic regionalization. In The Earth’s heterogeneous mantle
(pp. 3–46). Springer.
Şcheaua, F. D. (2016). Theoretical approaches regarding the Venturi effect. Coefficient of Small Orifices,69.
Schull, T. J. (1988). Rift basins of interior Sudan: Petroleum exploration and discovery. AAPG Bulletin,72(10), 1128–1142. https://doi.org/10.
1306/703c9965‐1707‐11d7‐8645000102c1865d
Sobh, M., Ebbing, J., Mansi, A. H., Götze, H.‐J., Emry, E., & Abdelsalam, M. (2020). The lithospheric structure of the Saharan metacraton from 3‐
d integrated geophysical‐petrological modeling. Journal of Geophysical Research: Solid Earth,125(8), e2019JB018747. https://doi.org/10.
1029/2019jb018747
Stuart, G., Fairhead, J., Dorbath, L., & Dorbath, C. (1985). A seismic refraction study of the crustal structure associated with the Adamawa plateau
and Garoua rift, Cameroon, west Africa. Geophysical Journal International,81(1), 1–12. https://doi.org/10.1111/j.1365‐246x.1985.tb01346.x
Svartman Dias, A. E., Hayman, N. W., & Lavier, L. L. (2016). Thinning factor distributions viewed through numerical models of continental
extension. Tectonics,35(12), 3050–3069. https://doi.org/10.1002/2016tc004266
Szwillus, W., Afonso, J. C., Ebbing, J., & Mooney, W. D. (2019). Global crustal thickness and velocity structure from geostatistical analysis of
seismic data. Journal of Geophysical Research: Solid Earth,124(2), 1626–1652. https://doi.org/10.1029/2018jb016593
Tokam, A.‐P. K., Tabod, C. T., Nyblade, A. A., Julia, J., Wiens, D. A., & Pasyanos, M. E. (2010). Structure of the crust beneath Cameroon, west
Africa, from the joint inversion of Rayleigh wave group velocities and receiver functions. Geophysical Journal International,183(2), 1061–
1076. https://doi.org/10.1111/j.1365‐246x.2010.04776.x
Uieda, L., Barbosa, V. C., & Braitenberg, C. (2016). Tesseroids: Forward‐modeling gravitational fields in spherical coordinates. Geophysics,
81(5), F41–F48. https://doi.org/10.1190/geo2015‐0204.1
Weatherill, G., & Burton, P. W. (2009). Delineation of shallow seismic source zones using k‐means cluster analysis, with application to the
Aegean region. Geophysical Journal International,176(2), 565–588. https://doi.org/10.1111/j.1365‐246x.2008.03997.x
Zingerle, P., Pail, R., Gruber, T., & Oikonomidou, X. (2020). The combined global gravity field model XGM2019e. Journal of Geodesy,94(7), 1–
12. https://doi.org/10.1007/s00190‐020‐01398‐0
Journal of Geophysical Research: Solid Earth
10.1029/2024JB029226
FOSSO TÉGUIA M ET AL. 23 of 23
