This study was carried out in the Lom series in Cameroon, at the border with
Central African Republic, located between the latitudes 5
There is a renewed interest in magnetic methods for solid mineral, hydrocarbon, and geological research. During data interpretation, the first crucial step is the removal of the effect of deep-seated structures from the observed total magnetic field to enhance shallow body signatures (Ndougsa et al., 2013). These shallow bodies in mining exploration are generally associated with mineral substances which have magnetic properties (Ndougsa et al., 2013). In our study, magnetic fabrics are signalled by Kankeu et al. (2009). The second step is mapping the causative body's edges, which is fundamental to the use of potential field data for geological mapping. The edge detection techniques are used to distinguish between different sizes and different depths of the geological discontinuities (Oruç et al., 2011). There have been several methods proposed to help normalize the magnetic signatures in images. Cordell and Grauch (1985) have suggested a method to locate horizontal extents of the sources from the maxima of horizontal gradient of the pseudo-gravity computed from the magnetic anomalies. Verduzco et al. (2004) discussed the use of tilt derivatives from gravity or magnetic field anomaly maps using the horizontal gradient magnitude of the tilt derivative as an edge detector for vertical contacts.
Salem et al. (2008) developed a new interpretation method for gridded magnetic data based on the tilt derivative, without specifying prior information about the nature of the source. In this article, we have used the approach of Salem et al. (2007) for the location of vertical contacts and source depth. In addition, because the identified source has a non-uniform volume from the roof to the bottom, we examine how this volume varies with depth by using upward continuation of magnetic anomaly.
The following structural domains can be distinguished in the Pan-African
Belt north of the Congo craton (Toteu et al., 2004; Fig. 1):
a pre-collisional stage that includes the emplacement of pre-tectonic
calc-alkaline granitoids (e.g. at 660–670 Ma); a syn-collisional stage inducing crustal thickening and delamination of the
subcrustal lithospheric mantle and comprising D1 and D2 deformations and
S-type granitoids (640–610 Ma; Toteu et al., 2004); a post-collisional stage associated with D3 deformation (nappe and wrench)
concomitant with the exhumation of granulites, development of D4 shear zones,
and emplacement of late-tectonic calc-alkaline to sub-alkaline granitoids
(600–570 Ma).
The Pan-African formations of Cameroon belong to the mobile zone of central
Africa (Bessoles and Trompette, 1980), also known as the Oubanguide chain
(Poidevin, 1985). It is attached to the east to Pan-African formations of the
Mozambican belt of sub-meridian orientation. To the west, it extends to the
north of Brazil by the Sergipe range. Two large dextral mylonitic shear
zones, the Sanaga Fault (Dumont, 1986) and the central Cameroon shear zone,
cross Cameroon from the north-east to the south-west. These major shears belong to the
Oubanguides setback zone (Rolin, 1995), which runs continually from the
Gulf of Guinea to the Gulf of Aden (Cornacchia and Dars, 1983). Geologically,
the Pan-African mobile chain is composed of granites, schists, micaschists,
and migmatites (Poidevin, 1985).
Geologic map of Cameroon, modified from Kankeu et al. (2009) and available in the public domain, showing major lithotectonic units: ASZ, Adamaoua shear zone; CCSZ, central Cameroon shear zone; TBSZ, Tcholliré-Banyo shear zone. The location of the study area is marked by a box and shown in detail in Fig. 2.
The study area is in eastern Cameroon; it is bounded by northern latitudes
5
Geological map of the study area (Gazel and Gerard, 1954, modified as a document available in the public domain). In the centre we have the Lom series marked by its greenschist facies. We can also perceive in red the tectonic lines that cross the study area.
The orography and hydrographic network is structurally guided (Kouske, 2006), sub-divided into three major morphological units. The high-altitude unit (800–1092 m) which is a vast peneplain enamelled by interfluves with multiple vertices of alignment oriented NW–SE, N–S, and NE–SW. The low-altitude unit (652–760 m) is a large flat-bottomed depression, in the centre of which is a U-shaped valley, oriented NE–SW, within which flows the Lom river and the intermediate unit (760–860 m), which corresponds to a long NE–SW-oriented cliff connecting the high-altitude unit to that of low altitude. In its northern part, this unit has an E–W orientation.
Seismic anisotropy in Cameroon has been studied by Koch et al. (2012)
through analysis of SKS splitting, which allowed four regions of
distinct anisotropy to be identified: moderately strong NE–SW-oriented fast polarization
directions (
Gravity studies were carried out, and Tadjou et al. (2004) identified many structures like contacts, dykes, fractures, and faults in the transition zone between the Congo Craton and the Pan-African Belt in central Africa. Shandini et al. (2011) found evidence in the northern margin of the Congo Craton of a deep structure, which corresponds to a classical model of the collision suture of the West African Craton and Pan-African Belt.
Owono et al. (2019) used 2.75D modelling of aeromagnetic data in Bertoua and showed intrusive bodies composed of gneiss and porphyroid granite and some domes with their roof situated at various depths not exceeding 1800 m from the surface. The structural map of the study area shows the trending of the structural features observed, namely, NE–SW, NW–SE, ENE–WSW, and WNW–ESE, while the E–W and N–S are secondary orientations of the observed tectonic evidence.
The aeromagnetic data were collected in Cameroon by Survair Limited through the Cameroon–Canada cooperation framework in the 1970s. Data were collected along N–S flight lines at 750 m spacing, with a flying height of 235 m; the measurements involved a magnetometer with a sensitivity of 0.5 nT (Paterson et al., 1976). Aeromagnetic anomaly maps have been digitized using the geographical information system software (Mapinfo Pro. 16.0) and interpolated on a grid with 750 m cells. The estimate error introduced is 0.28 mm, which is usually considered to be the distinctive capacity of human vision (Achilleos, 2010). Gridding and processing were done with Geosoft v8.4 software. The IGRF-70 reference field values were removed from the observed magnetic data as stated by Reeves (2005).
The upward continuation computes the magnetic field that would have been measured further away from the source, with is the smoothing operation. The upward continuation was proposed by Henderson and Zietz (1949) and described by Blakely (1996). In this study it helps us to easily visualize the effects of the deep sources and to remove their effect.
The tilt angle (Miller and Singh 1994; Verduzco et al., 2004; Salem et al.,
2007) is defined by Eq. (1) below for a potential field anomaly
In 2007, Salem et al. (2007) extended the method to the determination of depth to
source by relating the depth
The tilt-angle operator can be used for mapping geological structures
because it permits their contacts and their shapes to be located and delimited (Miller and Singh, 1994). By coupling it to the extension upward, it becomes
more interesting because one obtains the lateral extension of body but also
of depth and therefore its three-dimensional shape. Salem et al. (2007)
proposed the use of tilt angle for the localization of vertical contacts, knowing that the upward continuation operator can attenuate short
wavelengths and allow long wavelengths to be visualized (Henderson and Zietz,
1949). We can therefore use it for a better visualization of the behaviour of
contacts with depth. Thus, we have
generated the total magnetic intensity (TMI) maps reduced to the Equator and then applied upward
continuation for 1 and 2 km, computed the position of the contact considered as vertical of these
different three maps using Salem et al. (2007), and finally superimposed the different contact maps obtained to evaluate the
continuity of the sources. This technique is used in the qualitative
analysis for tilt-derivative results.
This method was introduced by Thompson (1982) based on the Euler
homogeneity equation to solve for the source depths for profile data. Reid
et al. (1990) extended the operator to gridded data by using Eq. (3):
Structural index for magnetic sources of different geometries.
A particularly useful variation on the 2D model, which removes the restriction of infinite strike length and is easier to define than the more complex 3D model, is a model with constant cross section extending over a finite strike length. This is known as a 2.5D model. When the source can have different strike extents on either side of the modelled profile, or the strike or plunge of the body is not perpendicular to the profile, this is called a 2.75D model.
The 2.75D model represents the subsurface as a series of polygonal prisms
with horizontal axes (
After interpolation, data have been reduced to the Equator using the Fourier
transform performed with inclination
The magnetic field over the Bétaré-Oya area has a complex magnetic pattern (Fig. 3). For better characterization of the geological structures, we subdivided the area into different units.
The major observable singularity is in the centre where a large anomaly about 5 km wide and up to 100 nT is observed. It is oriented NE–SW along the major tectonic feature in this area, namely the tectonic line of the Sanaga (Fig. 1). Comparing with the geological map in Fig. 2, this signal is mainly due to volcano-clastic schists (with gold deposit), also called Lom schists, associated with conglomeratic quartzites with intrusions of granitoids (Kankeu et al., 2009). Hence, the presence of the anomalies with similar signatures could be related to the circulation of hydrothermal fluids rich in magnetic minerals along the Bétaré-Oya shear zone (BOSZ).
In the north-eastern part of Bétaré-Oya, particularly around Badzéré, two heterogeneous anomalies are observed. They are in the south of the area at Ndokayo, with very long wavelength of about 22 km. Their amplitudes are quite high and reach 120 nT. They are aligned with one of the major foliations in this area trending E–W. The shape and amplitude of these anomalies suggest high magnetization of the causative bodies, such as igneous granitoids known to be in this area.
In Mararaba and Kassa, there is a large magnetic anomaly (Fig. 3). It is characterized by a long wavelength with variable amplitude reaching 150 nT; its approximate direction is ENE–WSW. We can also observe anomalies of 100 and 20 nT intensity; elongated shapes; circular and semi-circular shapes; and short wavelengths oriented ENE–WSW, NW–SE, and NE–SW corresponding to structural directions in the study area (Kankeu et al., 2009; Nih-Fon et al., 2012).
Total magnetic intensity (TMI) map reduced to the Equator.
In the north-western part of Garoua-Boulai, a heterogeneous anomaly with irregular shapes and a very long wavelength of about 22 km has been observed. Its amplitude is quite high and reaches 177 nT. Its approximate direction is ENE–WSW. It is probably associated with the meta-volcanic outcrops of the meta-lava within the schistous Lom series (Regnoult, 1986).
The lowest magnetic intensities are recorded in the north-west near
Djaororo-Bali, where anomalies with amplitudes down to
The residual map is obtained by subtracting the total magnetic field map
reduced to the Equator to the regional map. The determination of the optimum
regional anomaly map for the study area relies on the method of Zeng (1989).
This method consists of determining a suitable altitude for upward
continuation in the study area. The extrema of each altitude of upward
continuation are then counted (Table 2). These are points where the gradient
is null. Further, a graph of extrema versus altitudes of upward continuation
is plotted (Fig. 4). Finally, the suitable altitude (
Maxima and altitudes of upward continuation.
Number of extrema versus upward continuation height. From
Tilt angle on residual map.
The generated tilt angle map (Fig. 5) represents possible lineaments of
the study area. On this map it can clearly be seen that the signal is
uniformly distributed in
To characterize information, we were interested in the peaks of anomalies derived from the tilt-angle derivative (Fig. 5). We counted 111 lineaments, among which 45 have lengths varying between 2.5 and 10.8 km, 37 are minor lineaments varying between 1.2 and 2.3 km, and 29 are major lineaments between 2.4 and 15.6 km. Five structural NE–SW families are observed: ENE–WSW, E–W, NW–SE, N–S, and the major structural direction NE–SW (Fig. 6).
Structural map of the study area.
Major faults map superimposed on tilt-angle contours. On this map we observe the major regional deformation axes (NE–SW, E–W, ENE–WSW, and NW–SE) as well as the associated faults (F1 to F29).
The longest faults are present at the eastern edge of the Lom series with lengths of more than 15 km (F1, F3, F7). To the west we also note the NE–SW F4 fault with more than 10 km length, which marks the limit of the Lom series (Fig. 7). The most remarkable is the change of direction of compression or deformation axes. The E–W events marked by the faults F15, F16, F19, F25, and F26 at the eastern edge of the Lom and by the faults F5, F9, and F10 in the west seem to have been taken up by the tectonic accidents F1, F2, F3, F4, F7, F8, F12, and F23 punctuated by the BOSZ. The same phenomenon occurs in the extreme west of the study area around Dang Haoussa and Mararaba with the ENE–WSW (F13, F14, F18, F20, F21, F22) and NW–SE (F27, F29) accidents, respectively. These discrepancies suggest the passage of shear faults. The curvature (type II) structures corresponding to foliations induce most of the major fault network present in the Bétaré-Oya area. In order to confirm the results obtained by the tilt derivative, we apply the Euler deconvolution method.
By superposing the zero contours of tilt angle of the residual map, we obtain Fig. 8, which has no perfect superimposition of sources on the previous ones and hence assumes the heterogeneity of the basement and the existence of movements that affected the subsurface formations. Deep crustal tightening of volcano-clastic rocks in the vicinity of Bétaré-Oya confirms that the site is affected by shear tectonics (Soba, 1989), causing deep and shallow faults. This is shown by the contact between the granito-gneissic rocks and the Lom schists (Fig. 2). These contours delimit the edges of the magnetic source, so their superposition in depth allows an idea to develop about the disposition, the extent, the dip, and the shape of the geological sources responsible for the magnetic anomalies observed.
By applying the principles mentioned in Sect. 3.2.2, we observed the following from
the obtained map (Fig. 8 below):
They are not identical, which could mean that the contacts situated near the surface could be masked by those located at the subsurface or in
depth. There are some vertical contacts that narrowed with depth. This could
be interpreted as a sign of crustal thinning of the source of the anomaly
with depth. In some places, a lateral displacement of the contact is identified.
It could suggest here, a dip of the source in the concerned direction.
For example, the east of Ndokayo, Kassa, and the south-east of Mborguene,
several structures lose extension in depth, taking the form of a basic cone
of revolution located on the surface (interrupted circle).
The presence of this regional-scale fold system, which controls all movements in the area (BOSZ), suggests an interconnection of crustal geological structures by lines of faults and foliations. Hence the structural elements highlighted in this study (folds, faults, dykes, etc.) globally belong to Pan-African tectonics.
Superposition of contours (
The tilt-angle operator makes it easy to determine the depth of the vertical
contacts (Salem et al., 2007) by estimating the distance between the
zero-angle contours and those corresponding to the values
Euler solutions allowed us to verify the position of the contacts obtained by the tilt-angle method as well as their depth.
The superposition of the structural map with Euler solutions allowed us to delimit deep and superficial faults, dykes, and veins; to delineate tectonic lines established by previous geological studies (Gazel and Gerard, 1954); and to compare with results from the tilt-angle method (Fig. 10).
Euler solution (
Euler solution (
On the Euler solutions map we have perfectly distinguished the limits of the intrusive bodies and the deeper faults. On these maps, we observe the five main directions of structures, namely NE–SW, ENE–WSW, E–W, NW–SE, and N–S (Fig. 7). In addition, the vertical contacts are clearly visible on the Euler solutions map and extend over 15 km in length.
The deepest accidents are mainly NE–SW to E–W with depths of over 3500 m and are well located at the eastern limits in the Lom series the Badzéré gneiss contact zone, and around the East fault of Bétaré-Oya. In the south-west of the map, at Mararaba, Euler solutions allow NW–SE faults to be approximately detected that were the result of the highlighted tectonic line (Fig. 9) and whose depths are estimated at 3000 m. We obtain depths ranging from 0.5 to 3.6 km. Figure 10 clearly shows tectonic directions which dominate all subsurface movements of the study area and their depths ranging from 1 to 3.4 km.
Main faults of Lom series. This summary table is obtained after comparing the results from the Euler deconvolution method and the tilt derivative.
This profile (P1) extends 48.8 km NW–SE through Badzere and Mborguene. It crosses six geological formations from NW to SE, namely porphyroid granite, granite
with biotite, gneiss embrechites, granite of anatexis, schists, biotite, and
muscovite gneiss (Fig. 3). The strongest anomalies are localized in the NW
of the profile with an intensity of 177 nT. The basement obtained is made up
of granite anataxis which are old magmatic rocks forming the old basement
complex and put in place during the first half of the Precambrian. Its
maximum depth is
2.75D model obtained from profile P1.
Profile 2 (P2) extends 46 km along the NW–SE direction through
Bétaré-Oya and Kissi. It crosses five geological formations: biotite
leptinites gneiss, quartzite with muscovite schists, schists, biotite and
muscovite gneiss, and alkaline granite (Fig. 3). The lowest anomalies are
localized in the NW of the profile with an intensity of
2.75D model obtained from profile P2.
The geological synthesis of Cameroon allows us to have a lithostratigraphic sketch of the Lom Formation. Recently, the near-surface work of Mboudou et al. (2017) at Bétaré-Oya proposed the lithological model with topsoil, saprolites, sandy layer, conglomeritic sand, and schist formations.
On our model from profile 2 that passes through the locality of Bétaré-Oya, we observe that the first layers of rocks encountered are well below the topography, which is explained by the fact that the method used allows us to highlight the structuring of deep formations. This would have the effect of hiding the superficial (sediments) and hence the observed shift. Thus, the first formation detected on our models at Bétaré-Oya is schist. We can therefore complete this lithological model with the formations of the Pan-African basement highlighted by our geophysical methods (Fig. 13) and propose the litho-stratigraphic model updated below (Table 4). Crustal formations in our model are in accordance with those obtained by Benkhelil et al. (2002) from seismic data from south Cameroon, and the summary above is in accordance with the geological study of Mboudou et al. (2017).
Nature of formations.
Improved sketch of the lithologic profile of Bétaré-Oya basin. The scale map is 2
This model confirms the granite-gneiss nature of the Pan-African base.
The major faults highlighted in this work controlled by the BOSZ belong in fact to a wider network of faults found on the Pan-African and which would extend to the São Francisco Craton (SFC) by the central Cameroon shear zone (CCSZ). Indeed, the work of Toteu et al. (2004) suggests that the Reghane shear zone, which during the whole Pan-African Belt evolution (650–580 Ma) only recorded dextral wrench movement, can be considered as a major boundary separating the mobile domain in two (Fig. 14) – a western part where the tectonics is controlled by the motion of the WAC and an eastern part controlled by the motion of the Congo craton.
The Pan-African mobile domain (PMD) between the West African Craton (WAC) and the São Francisco (SFC) and Congo (CC) cratons showing two sub-domains, west and east, separated by the Raghane shear zone (RSZ). Horizontal lines represent the Tcholliré–Banyo shear zone (TBSZ), central Cameroon shear zone (CCSZ), Sanaga fault (SF), and Bétaré-Oya shear zone (BOSZ). Small arrows correspond to stretching lineations and large arrows to movement directions of blocks during D3 (600–580 Ma). Modified from Toteu et al. (2004) (original document is available in the public domain).
The structural map obtained (Fig. 6) shows a great disparity in the distribution of lineaments which can be explained in part by the general tectonics of the area. Hence, the collision between the stable Archean craton in the south and one of the two Paleoproterozoic blocks in the north during the Pan-African orogeny of 700 Ma would have caused a flattening of the basement and intrusions in the old Precambrian basement, causing the major NE–SW-oriented lineaments related to the Lom schists. According to the Cameroon geological synthesis, these intrusions are identified as granitic batholiths placed during regional deformation D1 and D2.
On both sides of the Lom series, there are major NE–SW lineaments representing the bounding faults of the Lom series with the granite-gneiss rocks. The E–W, NE–SW, and N–S lineaments may represent major tectonic structures marking the change in the structural direction between the trans-Saharan (N–S) and the Oubanguides (E–W) chains.
At the local scale, the deformation D2 is characterized by L2 lineations representing here stretches of quartz minerals oriented E–W. The ENE–WSW-oriented lineaments appear to correlate with the mylonitic deformations occurring during the D3 phase while the ones trending NW–SE related to senestral and dextral recesses and represent fractures with or without lode flow. These structures are much more abundant near Mararaba and could be the target for future mining studies.
The geoelectrical study of Nih-Fon et al. (2012) in our study area identified NE–SW-oriented irregular anomaly zones. These correlate with the quartz veins known in the region and are aligned with the regional shear zone. The morphological units identified also present NW–SE, N–S, NE–SW, and E–W directions. In addition, Kouske (2006) reveals that the hydrographic network of the study area has two major directions, NE–SW and NW–SE, and it is dense and dendritic.
The P1 and P2 models obtained can be used as pseudo-3D imagery of the Lom basement. Previous geological studies indicate that the area was subject to intense metamorphic activity during the Neoproterozoic that has resulted in schist formation (Coyne and Bellier, 2010). The contact between this schistous series and the gneissic and granitic rocks of the basement resulted in multiple fractures and faults (Gazel and Gerard, 1954; Soba, 1989). The litho-stratigraphic sketch proposed by our models, derived from the magnetic profiles and work of Mboudou et al. (2017), is consistent with previous geological work that asserts that the Pan-African basement would be made up of migmatites and granitic to ortho-gneissic and biotite-rich rocks (Poidevin, 1985; Gazel and Gerard, 1954; Kouske, 2006; Ngako et al., 2003; Koch et al., 2012; Pepogo et al., 2019; Owono et al., 2019; Eno Belinga, 1984).
From the mining point of view, the artisanal gold indices are places located near the Lom and Pangar rivers (Nih-Fon et al., 2012). These alluviums correlate with NE–SW trends in our structural map. Since the structures in our study area are structurally guided, it can be concluded that the alluvial deposits observed and exploited by residents are some signs that have been leached and transported by the waterways. Overall, the geological structures obtained from the data processing correspond to the ductile-brittle structures such as shear zones and faults. These structures constitute pathways for both mineralizing fluids and groundwater. Since several gold mines exist in the Bétaré-Oya area, the new mapping approach could be an important guide for the identification of the structures that control the gold mineralization in the area.
In this work, some new analysis techniques were applied on aeromagnetic data to delineate the sub-surface structures. The results obtained highlight the axes of compression, folding, and shearing as well as mylonitic veins (veins are at the outcrop's scale) several kilometres long and oriented NE–SW. The regional and local structural settings of the area are characterized by major faults and other structural elements mainly striking in the NE–SW, NW–SE, ENE–WSW, N–S, and E–W directions. A major trend in the NE–SW direction represents the dominant tectonic trend which is the prolongation of the central Cameroon shear zone (CCSZ) in the study area. Several folds and faults evidenced by this study correlate with past studies while others are inferences. The depths of major accidents in the area have been estimated between 1.2 and 3.6 km, and the NE–SW structures on our structural map are proposed here for possible gold exploration. The models from the P1 and P2 profiles have enabled us to propose a structuration of the superficial crust of the Lom that highlights the main rocks and intrusions responsible for the observed anomalies (porphyroid granite, garnet gneiss, syenites, micaschists, graphite, and garnet gneiss), identify deep and shallow fractures, their depths, and propose a lithostratigraphic model in agreement with the previous works. Finally, we note that the tilt angle coupled to the upward continuation is an interesting tool for 2.75D modelling.
Despite the results obtained, some limits are observed concerning the
application of the tilt derivative on data RTE, as cited below:
With the data RTE, we used the hypothesis of a magnetically induced anomaly, which
has the same direction as the geomagnetic field, given by the global
geomagnetic field models (IGRF), and the remanent part in the rock source of
the anomaly seems to be neglected. Satisfactory results using the tilt angle on magnetic data analysis and
modelling, especially in the determination of the vertical contacts in low-latitude areas, are constrained by the use of the data RTE. use of the Euler deconvolution to identify various potential contacts
including the vertical one and then compare them with those derived from the
tilt results, and care to correlate our results with the previous geological investigations
over and outside on the one hand, and geophysical studies over adjacent areas,
particularly the aeromagnetic investigations, on the other hand.
In our case, investigating a low-latitude area we engaged the following:
The data used to support the findings of this study are available from the corresponding author upon request.
CEN performed the data analyses, modelling, and preliminary interpretation including preparation of the paper in conjunction with all the co-authors. TNM designed the topic, gave the orientations for the investigation, and reviewed the quality of the models and related interpretation and the entire paper. MBBA defined the criteria and the physical parameters for the 2D3/4 modelling with the first author. SAMT and SPA worked on the review of quality and quantitative analyses of the maps and 2D3/4 models.
The authors declare that they have no conflict of interest.
The authors thank the reviewers for their valuable comments.
This paper was edited by Nicolas Gillet and reviewed by Bruno Gavazzi and one anonymous referee.