• Leandri Venteru

Mid- African Rift: Is there a geographical/spatial control on the nature of volcanism in Cameroon?

Essay

Volcanism in Cameroon occurs mainly on the Cameroon Volcanic Line (CVL), with the largest and most active volcano being Mount Cameroon (Fitton, 1980). The CVL, representing alkaline volcanoes, shows an easy recognisable Y-shape, with volcanic centres in both the oceanic and continental sectors of the African plate (Fitton & Dunlop, 1985). Consensus regarding the driving forces of these volcanoes has yet to be reached. However, a wide variety of theories and hypotheses addressing the origin and nature of the CVL have been published over the last few decades.

The CVL is a 1600km long magmatic sequence, stretching from Pagalu Island in the Gulf of Guinea (Atlantic Ocean) to Lank Chad on the main land of Cameroon on the western African continent (Déruelle, Ngounouno & Demaiffe, 2007). Geographically the CVL is situated almost parallel to the Benue Trough, and follows the length of the Ngaoundéré Fault (Ballentine, Lee & Halliday, 1997). The northern limb of the CVL overlaps with the Chad rift, which marks the northern rift of the Benue Trough (Fitton, 1980). Although volcanism is still active along the CVL is it not a neogene feature and has been active since the beginning of the Tertiary period. The rocks from the CVL are classified as mainly alkaline, intermediate and felsic. The felsic and basaltic lavas increase inlands towards the branches of the CVL in a volumetric ratio (Déruelle, Ngounouno & Demaiffe, 2007). One of Africa’s largest volcanoes, Mount Cameroon, is a volcanic horst, with an approximate height of 4075m. The volcano of Mount Cameroon is the most active volcano on the CVL. The latest eruption of Mount Cameroon was in the year 2000 (Herrero-Bervera et al. 2004).

Scientists tried various methods over the years in an attempt to clarify the origin and dynamics of the CVL. During the 1980’s Fitton (1980) proposed a simple model to explain the geological features of the CVL. Fitton (1980) noted that although there is no evidence for rift faulting there are signs of regional uplift of the basin. The continental sector of the CVL is characterized by strato-volcanoes (Mount Cameroon), central volcanic massifs and even calderas in some areas. Collapsed plains filled by sediments, single magnetic volcanoes and flood basalts ( Ngaoundéré Plateau) are also prominent on the continental sector (Nkouathio et al. 2008). The volcanism is not influenced by fractures, which existed before the volcanism in the basement (Fitton, 1980). The origin of the CVL is explained by hand of the striking relationship between the CVL and the Benue Trough features. Fitton (1980) suggested that the CVL and the Benue Trough are superimposed by rotating one feature by 7° in relation to the other, about an axis. The axis is described to run with a north eastern strike into the country of Sudan. The reason for the relative rotation of the CVL and the Benue Trough is speculated to be as a result of the clockwise rotation of Africa (ca. 80Ma – 65Ma) (Fitton, 1980). This clockwise rotation might provide evidence for three features. Firstly, on the origin, the size and Y-shaped geometry of the CVL, secondly the rotation might support reason for the absence of recent volcanism in the Benue Trough. Thirdly, the rotational theory might provide evidence for the folding of the Benue Trough sediments. There were some speculations that the volcanism presently active in the CVL was once active in the Benue Trough. If the volcanic line had migrated from the Benue Though to its current position a systematic migration of a hot-spot is implied. Ngako et al. (2006) used remote sensing data as proposed by Moreau et al. (1987) to suggest that the relationship between stress regimes and intraplate alkaline magmatism in the CVL is controlled by lithospheric structures.

Ubangoh et al. (2005) argued that the presence of numerous recent volcanic cones and craters along the CVL is an indication of a strong geothermal gradient beneath the volcanic line, although there are no data that proves an increased supply of heat. By using a geochemical study on the basaltic rocks of the CVL and a K-Ar dating programme, Fitton and Dunlop (1985) searched for evidence to prove a systematic migration of the CVL volcanoes. Since the CVL is situated in such a unique geological setting, Fitton and Dunlop (1985) reasoned that they can determine the source of the CVL magmas by comparing the isotopic ratios of alkali basalts on the oceanic sector with the isotopic data of the alkali basalts on the continental sector of the volcanic line. Fitton and Dunlop have found that both the major element compositions as well as the isotopic ratios are indistinguishable when comparing the oceanic and continental sectors. What the K-Ar data did prove was that multiple volcanoes were active along the line in more or less the same time period. When Ballentine, Lee and Halliday (1997) compared the ages of the oceanic sector volcanoes, they found that there is a consistency between the rotation of the Cretaceous African plate (Fitton, 1980) and the ages of the earlier exposed rocks on the islands. The ages of the early exposed lavas decrease from the continental side towards the ocean, placing Principe at 31Ma and Pagalu at 4.8Ma. The systematic age decrease of the islands support the mantle plume model for the oceanic sector of the CVL, but does not include the continental sector. Other causes of volcanism, such as rift faulting and pre-existing basement fractures were at first discarded due to failure to collect supportive evidence. The volcanoes on the ocean floor showed no changes due to the faults they passes through, thus one can assume that the source of these volcanoes are from mantle processes and is not affected by structures occurring in the crust (Fitton, 1980).Research based on the age and chemistry of the CVL basalts, indicated that the CVL is a young rift associated with the Benue Trough rather than a hot-spot trail.

The depth at which a volcano is tapping its magma can be determined by the radioactive decay of helium isotopes in crustal rock (Aka, et al. 2003). Helium serves as a geochemical tracer that could be used to investigate the alteration of magma over a period of time. It is generally accepted that ocean island basalts (OIB) show a wide range of ³He/⁴He (Zindler & Hart, 1986) and that ³He/⁴He ratios from mid ocean ridge basalts (MORB) are uniform (Graham et al. 1992a). The ³He/⁴He ratio of the CVL ranges from 3.05Ra to 8.31Ra. Hotspots in general have much higher ³He/⁴He ratios (Hilton et al. 1999). Aka et al. (2003) was the first to identify that the distribution of the ³He/⁴He ratios on the CVL could be due to spatial control, and suggested that the helium isotopic distribution an essential characteristic of the CVL is. These findings are argued as prove that the CVL volcanoes doesn’t tap their material directly from hotspots. Aka et al. (2003) is further backed up by work Fitton and Dunlop (1985) did on K-Ar dating and the consistency to the Pb-isotope model proposed by Halliday et al. (1990).The above mentioned data and the long-lasting volcanism of the CVL volcanoes supports the conclusion reached by Fitton and Dunlop (1985) that the magma driving the volcanoes on the continental sector could not originate from a deep mantle source. Therefore the CVL shows no evidence of a mantle plume in the continental sector, but rather signatures convection of the upper mantle (Fitton & Dunlop, 1985).

²⁰⁶Pb/²⁰⁴Pb and ²⁰⁸Pb/²⁰⁴Pb isotope ratios showed a slight decrease to either sides of the volcanic line from the centred volcano, Mt. Cameroon, which lies on the lithospheric continental-oceanic boundary (COB) (Ballentine, Lee & Halliday, 1997). Basalts from the COB volcanoes, such as Mount Cameroon and Mount Etinde are more radiogenic with ²⁰⁶Pb/²⁰⁴Pb ratios of approximately 20.52 and ²⁰⁸Pb/²⁰⁴Pb ratios of 40.34 (Déruelle, Ngounouno & Demaiffe, 2007). The Pb ratios of the COB volcanoes show a strong contrast to the lower radiogenic ratios of the island volcanoes. Basalts from the island located far south from the COB, Pagalu Island, have the lowest ²⁰⁶Pb/²⁰⁴Pb ratio of 19.01 and a ²⁰⁸Pb/²⁰⁴Pb ratio of 38.83. The Pb isotope data proves that there occurred no consistent major change in the magma over a period of time. The Nd and Sr isotopes of the basalts from both the continental and oceanic regions are similar in composition. The resemblance of the geochemical and isotopic data between oceanic and continental basalts provide evidence that the source and composition of the magma was not influenced by the continental crust, and therefore, the magma does not originate from the lithosphere (Déruelle, Ngounouno & Demaiffe, 2007). The magma, as speculated by Ballentine, Lee and Halliday (1997), originate from a sub-lithospheric source, because of the similar compositions of the oceanic and continental magmas. Nkouathio et al. (2008) argued that since the lavas of the entire CVL have a shallow asthensopheric mantle source a depleted MORB mantle and a metasomatosed mantle, the alkaline lavas are consistent with the asthensopheric upwelling of a hot-spot. Halliday et al. (1990) argued that the isotopic decreases could infer re-melting of an enriched mantle plume located beneath the COB. Halliday further argues that the plume possibly shifted (from its position underneath the Benue Trough) to its current location during the continental breakup between the African and South American plates as well as the rotation of the cooler lithosphere relative to the asthenosphere (Ballentine, Lee & Halliday, 1997). Ballentine, Lee and Halliday (1997) made it clear that although the isotopic data supports the mantle plume theory, the mantle plume is only a part of the explanation of a more complex volcanic setting.

Detailed isotopic data demonstrated that many theories proposed in the past were too inconsistent to explain the origin of the CVL. The data includes many classical theories, however only three will be mentioned. Firstly, the reactivation of the Ngaoundere fault which in turn caused the volcanism on the CVL (Moreau et al. 1987). A second classical theory involves membrane tectonics which produced the extensional CVL feature (Freeth, 1979). Thirdly, it is suggested that the various active volcanic centres are inconsistent with the ‘textbook’ model of hotspots (Fitton & Dunlop, 1985), (Ballentine, Lee & Halliday, 1997).Keeping the above mentioned theories in mind, Ballentine, Lee and Halliday (1997) used a ‘technique of inductively coupled plasma magnetic sector multiple collector mass spectrometry’ (Ballentine, Lee & Halliday, 1997 pp 111)to acquire new data. From the data Ballentine, Lee and Halliday (1997) concluded that the CVL does not show a mentionable difference in Hf and Nd isotopic compositions in neither the continental margin nor in the oceanic sectors. However, a very significant difference in Pb and Sr isotopic compositions were detected (Ballentine, Lee & Halliday, 1997). The consistent Hf and Nd isotopic compositions in the ocean island basalts (OIB) are indicative of a similar behaviour of the Sm/Nd and Lu/Hf parent-daughter ratios in the source of the OIB (Ballentine, Lee & Halliday, 1997). Further, the continental sector shows a similar Hf-isotopic make-up than the oceanic sector. Ballentine, Lee and Halliday (1997), argued that the distinct variability in the Pb isotopic composition can be explained by the recycling of the oceanic crust, since Pb and Sr have increased mobility in the oceanic lithosphere.

Kamgang et al. (2008) used a more localized approach, by investigating the geochemical isotopic data of the Bamenda Mountains in Cameroon. Since the previous isotopic studies performed by Ballentine, Lee and Halliday (1997), focussed on the entire CVL, Kamgang et al. (2008) proposed a ‘zoomed in’ investigation by focussing on one area of the CVL and try to explain the origin of the magma in that area. Kamgang et al. (2008) performed geochemical and geochronological studies of the mafic rocks from the Bamenda Mountains. His aim was to gain a better perspective of the composition of the CVL magma source. K-Ar dating performed on older, crust contaminated, samples of the CVL showed no systematic variation over time. However, insight on the chemical composition of the mantle source of magma is provided by the high Eu, Sr and Ba concentrations (Kamgang et al. 2008). The high concentrations are not a result of crustal contamination, because these concentrations do not correlate to the Sr and Nd isotopic compostitions or the MgO contents or the La/Nb ratios of the rocks. Kamgang et al. (2008) suggested that these high concentrations rather have to do with the composition of the magma origin. The major element composition shows consistency with fractional crystallization processes. Sr and Nd isotopic values of the Bamenda Mountains coincide with the isotopic data published by Ballentine, Lee and Halliday in 1997. Kamgang et al. (2008) pointed out that the positive correlations observed between the Pb and Nd isotopic ratios and the negative Pb/ Sr isotopic correlation indicate either contamination by the continental crust (Ngounouno, Déruelle & Demaiffe, 2000) or the effect of an enriched lithospheric mantle (Rankenburg, Lassiter & Brey, 2005)( Kamgang et al,2008).

Structural causes of volcanism have been discarded due to failure to prove structural influences with supportive evidence. There are three main structural units present in the Pan- African basement, namely shear zones, fold zones and thrust zones from which the shear zones indicating lithospheric faults stands out (Ngako et al. 2006).The volcanoes on the ocean floor show no changes due to the faults they passes through, thus one can assume that the source of these volcanoes are from mantle processes and is not affected by structures occurring in the crust (Fitton, 1980). It was also speculated that the CVL is rather a young rift associated with the Benue Trough. Nkouathio et al. (2008) suggested that the CVL is an alternating structure consisting of horsts and grabens. As a large shear zone, the CVL is structurally subdivided by a sequence of faults indicative of an alternating horst and graben system (Nkouathio et al. 2008).

Considering other tectonic models, Ngako et al. (2006) suggested that the magmatism of the CVL is a complex interaction of multiple mantle plumes and fractures in the lithosphere. Ngako et al. (2006) further speculate that the complex interaction may cause the new magmatic complexes to be diagonally aligned. By expanding the study area form only the CVL to other large scale magmatic provinces on the African continental plate, Ngako et al. (2006) tries to illustrate a complex interaction between hotspots and Precambrian faults. The aim is to provide alternative explanations to classical hotspot models. Ngako et al. (2006) used three provinces, namely the Niger- Nigeria super province (Plaeozoic to Mesozoic era), the Benue Trough (Cretaceous period) and the Cameroon Volcanic Line (Cenozoic era to present) to prove a time-space migration of the West-Central African alkaline magmatism. The ages of rocks sampled, decrease from the northern Niger- Nigeria province southwards through the Benue Trough to the CVL. The decreasing age of the plutonic provinces may suggest an interaction between tectonics and a hotspot system. Ngako et al. (2006) noted that the crustal and elastic thickness of the lithosphere on the continental sector of the CVL is unusually thin. This lithospheric thinning is argued as a position of a hotspot. Nkouathio et al. (2008) suggested that the evolution of the CVL magmas could be explained by the pattern of tectonic stresses and volcanic structures. In contradiction, Milelli, Fourel and Jaupart (2012) suggested that the Y-shape of the CVL was unaffected by plate motions and therefore implies that the magma source is attached to the continent.

Geophysical surveys performed on the crust of the CVL concluded that differentiation between the Pan-African belts and neither the structural features nor the thickness of the crust is possible (Milelli, Fourel & Jaupart, 2012). The crust underneath the CVL is very thin (ca. 35-39km) and the failure to distinguish between the Pan-African belts and the CVL indicates that the long period of magmatism (for approximately 70 Myr) on the CVL, has not have significant effects on the crust in the CVL region (Tokam et al. 2010). The absence of a seismic velocity anomaly at depths greater than ~300km, confirm the presence of the uniform mantle transition zone (Reusch et al. 2010)(Milelli, Fourel & Jaupart, 2012). Milelli, Fourel and Jaupart (2012) explored the buoyancy of the lithospheric origin, because the lithosphere is made up of depleted mantle material, relative to the asthenosphere as a basis for their experiment on lithospheric instabilities. Since the lithosphere is buoyant on the asthenosphere, the system is in equilibrium. However, if the lithosphere would experience cooling from the top, a convective instability might be triggered (Milelli, Fourel & Jaupart, 2012).

The latest trend is that researchers consider the CVL as a hot line tapping its resources from a sub-lithospheric mantle through openings in the lithosphere (Déruelle, Ngounouno & Demaiffe, 2007). Crustal uplift in both continental and oceanic areas proposed by Meyers et al. (1998) contributes to the theory of lithospheric instability proved by Milelli, Fourel and Jaupart (2012) via laboratory experiments. Milelli, Fourel and Jaupart (2012) studied the lithospheric instability due to cooling from the crust downwards. The lithospheric instability is speculated to occur within the sub-continental lithospheric mantle. The laboratory experiments performed by Milelli, Fourel and Jaupart (2012) were based on the principle that materials with a viscosity that is independent of temperature will react the same when cooling from either the top or the bottom was induced. A ‘continents’ were created by using different viscosity fluids and fixing certain variables. These experiments, that’s been fully described and discussed by Milelli, focussed on the instabilities created at the continental margin. Milelli, Fourel and Jaupart (2012) documented that the buoyancy factor causes downwellings that eventually return to the surface. This cycle of downwellings and upwellings creates a stress regime where compressional stresses were identified at the downwellings and corresponding extentional stresses, creating rifting fractures, at the upwellings. Fracture patterns on the surface of the experimental ‘continent’ formed in a combination of hexagons and distorted squares (Milelli, Fourel & Jaupart, 2012). The conclusion of these experiments serve as proof that that continental lithospheric cooling can result in the geometrical Y-shape of the CVL, independent of the size and shape of the continental mass. Further proof is presented by Déruelle, Ngounouno and Demaiffe (2007) with supporting data on seismic and gravimetric surveys of the oceanic section of the CVL.