Journal of African Earth Sciences 100 (2014) 164–178

Contents lists available at ScienceDirect

Journal of African Earth Sciences

journal homepage: www.elsevier.com/locate/jafrearsci

Origin and evolution of Ngaye River alluvial sediments, Northern

Cameroon: Geochemical constraints
Paul-Désiré Ndjigui a,⇑, Anicet Beauvais b, Soureiyatou Fadil-Djenabou a, Jean-Paul Ambrosi b
a
Department of Earth Sciences, University of Yaoundé 1, P.O. Box: 812, Yaoundé, Cameroon
b
Aix-Marseille University (AMU), IRD, CNRS, CEREGE UM34, BP 80, 13545 Aix-en-Provence, Cedex 4, France

a r t i c l e i n f o a b s t r a c t

Article history: The origin of Ngaye River alluvial sediments and the evaluation of the weathering degree of their source
Received 30 May 2013 rocks are assessed using trace and rare-earth element geochemistry in three bulk sediments and their dif-
Received in revised form 1 June 2014 ferent size fractions (2000–200 lm, 200–50 lm, 50–2 lm and <2 lm). The alluvial sediments consist of
Accepted 8 June 2014
two sandy clay layers at the bottom and one sandy heavy clay layer at the top. Quartz and feldspars are
Available online 24 June 2014
the main minerals in the sand fractions while kaolinite and smectite are dominant in the finest sediments.
The relatively low Chemical Index of Alteration (CIA) indicates that the sediments and their potential source
Keywords:
rocks are moderately weathered. Highest trace element contents are observed in the fine sands, which are
Alluvial sediments
Trace elements
the richest in Zr, Th, U, Sc and REE. La, Ce and Nd are the most abundant REE in this fraction. The coarse frac-
Rare-earth elements (REE) tions are characterized by LREE-enrichment relative to HREE. The PAAS-normalized REE patterns exhibit
Weathering large positive Eu anomalies (Eu/Eu 3.1 to 3.9) in the coarse sand fraction of the sandy clay layers and
Granitoids strong negative Eu anomalies (Eu/Eu 0.35 to 0.70) in the two sand fractions of the sandy heavy clay layer.
Northern Cameroon Our results document the immaturity of the Ngaye River sediments, which derive mainly from the erosion
of moderately weathered granitoids of the surrounding reliefs, and in some extent from greenstones and/or
basic volcanics. The results also suggest an obvious dependence of trace and rare-earth element fraction-
ation on mineral sorting and weathering in the different grain-sized fractions of the alluvial sediments.
Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction profiles showed depletions in the upper ferruginous horizons and

enrichments in the lower saprolite layers (e.g., Braun et al., 1993).
Geochemistry of clastic sedimentary deposits usefully docu- Geochemistry of Kaveri River alluvial sediments in South India
ments the source rocks that fed them (McLennan et al., 1993). (Singh and Rajamani, 2001) revealed that REE concentrations
Clays of alluvial sediments are either inherited from weathered depend on the sedimentary particles size, the finest sediments
source rocks or neo-formed by post-deposit weathering or diage- being richer in REE than the coarse sand fraction and bulk
netic processes (Singh and Rajamani, 2001; Njoya et al., 2006; sediment. Skolkovitz (1988) showed that river-borne sediments
Ngon Ngon et al., 2009, 2012; Singh, 2009; Lawal and Abdullahi, are strongly depleted in HREE relative to shale. Concentrations of
2010; Nguetnkam et al., 2011). Notwithstanding the primary con- Al, Fe, Ti, Th, Sc, Co, Zr and REE in the finest sediments
trol of source rocks composition, the grain size and geochemistry of (50 < Ø < 2 mm) are commonly linked to the host primary minerals
sediments also depend on weathering and diagenetic processes including accessory minerals (Taylor and McLennan, 1985; Singh,
(Sawyer, 1986). 2009; Bhuiyan et al., 2011). It was shown that a positive Eu anom-
Trace element distribution including rare-earth elements (REE) aly occurred in the sand fractions, whereas the finest sediments
was previously studied in weathering profiles developed upon var- were rather characterized by a negative Eu anomaly. The positive
ious rocks under different climate conditions (e.g., Nesbitt, 1979; Eu anomaly was related to feldspars, which together with quartz,
Braun et al., 1990; Beauvais and Colin, 1993; Boulangé and Colin, are the main components of the coarse sand fraction (Singh and
1994; Ndjigui et al., 2008, 2013; Beauvais, 2009; Etame et al., Rajamani, 2001; Singh, 2009). Here, we constrain the origin of
2009). In Southern Cameroon, REE fractionated in weathering the alluvial sediments in the Ngaye River valley (Northern Camer-
oon) by the quantification and distribution of major, trace and
rare-earth elements in different grain-sized fractions of these
⇑ Corresponding author. Tel.: +237 9954 3774; fax: +237 2222 6262. sediments. We also discuss the potential links between the REE
E-mail address: [email protected] (P.-D. Ndjigui). behavior and the sorting and weathering degree of the sediments.

http://dx.doi.org/10.1016/j.jafrearsci.2014.06.005
1464-343X/Ó 2014 Elsevier Ltd. All rights reserved.
 P.-D. Ndjigui et al. / Journal of African Earth Sciences 100 (2014) 164–178 165

2. Geographical and geological setting composed of sedimentary and meta-igneous rocks that later
underwent medium-to-high grade Pan–African metamorphism
The Ngaye River watershed is a tributary of the Mbéré River (Toteu et al., 2001; Tchameni et al., 2006). The sedimentary forma-
draining the Cretaceous Trough in Northern Cameroon (Figs. 1 tions are constituted from the bottom to top by: (i) conglomerates
and 2a). The Mbéré Cretaceous Trough is deep and located from with rounded pebbles; (ii) fine-grained sand mixed with clay
the northern border of the Adamawa Plateau in Cameroon to the patches; (iii) basanite flow; (iv) marl and probably clays; and (v)
southern prolongation of the Yade Massif in Central African Repub- coarse conglomerates (e.g., Guiraudie, 1955; Lasserre, 1958;
lic (CAR). The investigated Ngaye alluvial deposits are located at Koch, 1958). The age of the upper part of this series is Middle
7°110 4000 N and 14°580 5000 E (Fig. 1a). The climate is semi-arid, with Cretaceous by correlation with the Lame series in Chad (Le
a mean annual rainfall of 970 mm, and a mean annual temperature Maréchal and Vincent, 1971). The basement of the Ngaye region
of 27 °C (Molua, 2006). The vegetation is savanna associated with is mainly constituted of granitoids (Fig. 1b).
dense forest galleries in the valleys (Letouzey, 1985). The topography of Ngaye region is characterized by high reliefs
The Adamawa-Yade region is characterized by a NE–SW with a maximum altitude of 1300 m in the NW and SW parts
elongated regional-scale plutonic complex which intrusion into a (Fig. 2a and b), which dominate vast alluvial plains (Dirasset
Paleoproterozoic basement, which is locally covered by Cretaceous et al., 2000) drained by the Vina and Mbéré Rivers (Fig. 1b). Two
deposits (Mbéré and Koum basins) and Cenozoic rocks such as topographic levels are prominent (800–1000 and 1000–1200 m,
andesitic basalts belonging to the Cameroon Volcanic Line Fig. 2a). The landscape of the Ngaye River watershed consists of a
(Guiraudie, 1955) Southwest of Ngaye (Fig. 1b). The Adamawa- succession of hills separated by small and narrow valleys with
Yade batholith consists of a great variety of more or less deformed eroded or gullied bell-shaped bottoms (Figs. 1b and 2a). The soils
rock-types of different ages. The lithologies include granites, dior- exposed on the interfluves are 10 m thick, mainly ferruginous,
ites, gabbros and syenites (Tchameni et al., 2006). The granites and consist of six horizons from the bottom to top: coarse and fine
were emplaced into an Eburnean (2.1 Ga) remobilized basement saprolites, two nodular horizons embedding an iron duricrust, and

15°00'E 15°30'E
NIGER
Ndjamena
8°00'N

8°00'N
Kano

CHAD

Bo
rde
NIGERIA
Abuja Fig. 1b

r
Touboro

Ecole Vogdjom Vina

ne
Logo
Vina
CAMEROON CAR Baibokoum
Ugara
Obogo

Malabo Douala be
Ye

Yaounde ré
bé
Divide M
7°30'N

7°30'N
m
a 250 km Li
a
nd
Bi

Dompta
MNT SRTM (90m) UTM projection

Elevation (m)

400 - 600 Ngaye ré

Bouloupou bé
M
600 - 800

800 - 1000 Sokorta Manga

Study site
Fig. 2a
Lim

1000 - 1200 Ngou

7°00'N

7°00'N

N
1200 - 1945 Bertoua
der
Bor

Rivers 20 km
b
15°00'E 15°30'E

Precambrian syntectonic granitoids

(calco-alcaline granites) Cretaceous formations (sandstones, marls, limestones)

Cambro-ordovician syn- and post tectonic granitoids

(alcaline granites) Basalts (andesites) Faults

Granitoids with greestones facies Scarp faults with mylonites

(gneiss and migmatites; leptynites, quartzites, amphibolites, pyroxenites)

Fig. 1. (a) Location of the study site and (b) combined geological and geomorphological map of this site and Ngaye region (geological map adapted from Le Maréchal, 1976).
166 P.-D. Ndjigui et al. / Journal of African Earth Sciences 100 (2014) 164–178

SW NE 1100

Ugara

Elevation, m
1060

Ngaye 1020
Ngaye

Ngaye 980
Ngaye River
940
b
0 1 2 3 4 5
ré Length, km
N bé
a 5 km M
Study site Sample location
0
Sandy heavy clay

0.2
upper

MD3 0.4
0.6
0.8
middle

1.0
MD2 1.2

Thickness, m
1.4
Sandy clay

1.6
1.8
lower

2.0
2.2
MD1
2.4
2.6
Gravelly bed

2.8
3.0
3.2
c 3.4

Fig. 2. (a) Geomorphological and drainage map of the Ngaye River watershed, and location of the topographic cross section (b) showing the Ngaye River incision, and the
investigated alluvial terrace. (c) Vertical section across the alluvial deposits of the Ngaye River valley with collected samples (white squares) in the different alluvial
sedimentary layers. For the legend of (a), see Fig. 1b.

the loose clayey horizon exposed at the ground surface (Brabant sieving. The size fractions above 50 lm were determined by dry
and Gavaud, 1985). The loose clayey horizon may sometimes sieving.
connect with the alluvial deposits in lowlands.
3.2. Analytical techniques

The size fractions and the bulk samples were processed for min-
3. Sampling and analytical techniques eralogical and chemical analyses at the Geoscience Laboratories
(Sudbury, Canada). Powders were first prepared in the Department
3.1. Sampling and separation techniques of Earth Sciences (University of Yaoundé 1, Cameroon). Heavy min-
erals (160–100 lm) were concentrated by density separation using
Three bulk sediment samples, one hundred kilograms each, heavy liquid (bromoform) and identified under a binocular micro-
were collected in each layer of a profile in the terrace of the Ngaye scope. About two hundred grains of heavy minerals were collected
River (Fig. 2c). Samples were split up homogeneously and prepared for each sample; data are expressed in weight percent. The mineral
for granulometric, mineralogical and geochemical analyses. Grain compositions were determined by X-ray Diffraction (XRD) analysis
size was measured according to the standard pipetting method according to two steps. Bulk and grain size sediment samples were
for the silt- and clay-sized fractions (Singh, 2009). After that, 20 g ran using PAN Analytical X’PERT PRO diffractometer in the Geosci-
of the bulk sediment samples were treated with cold 1 M HCl ence Laboratories at 40 kV and 45 mA. Oriented samples of the
and H2O2 to remove organic matter. Adding sodium hexameta- clay- and silt-sized fractions were characterized using a Philips
phosphate deflocculated the samples. An aliquot of the bulk sedi- instrument (PW1050/81, PW3710) in the Aix-Marseille University
ment below 2 mm grain-sized was separated in four fractions: (Aix-en-Provence, France). This analytical instrument is equipped
coarse sands (2000–200 lm); fine sands (200–50 lm); silts with a monochromator using a Co Ka radiation of 1.7854 Å over
(50–2 lm) and clays (<2 lm). Sands were separated using wet a range 2.5° to 35° 2h with a step size of 0.05° 2h/min at 40 kV
 P.-D. Ndjigui et al. / Journal of African Earth Sciences 100 (2014) 164–178 167

and 40 mA. The oriented sub-samples were prepared by mounting The geochemical results (major, trace and rare-earth elements)
the clay- and silt-sized fractions on glass slides. Clay mineralogy were compared to the average compositions of granitoids such as
was determined by applying different treatments on sub-samples: biotite ± muscovite granitoids (BMG), biotite granitoids (BG) and
untreated, ethylene glycol, hydrazine and heated (550–600 °C). The hornblende-biotite granitoids (HBG) (see Tchameni et al., 2006)
untreated, heated and ethylene glycol sub-samples were X-ray which could be the potential source-rocks of Ngaye River alluvial
analyzed for 30 min (2.5 s counting time), and the hydrazine sediments (see Fig. 1b).
sub-samples for 10 min (1 s counting time).
Samples for geochemical analysis were crushed using a jaw- 4. Results
crusher with steel plates, and pulverized in a ball mill made of
99.8% Al2O3. Powders were first heated at 105 °C in the presence 4.1. Granulometric and mineralogical characterization
of oxygen to drive off remaining volatile components and oxidized
Fe. The loss on ignition (LOI) was determined at 1000 °C on dry The Ngaye alluvial deposits consist of three layers (Fig. 2c)
(105 °C) samples. X-ray Fluorescence (XRF) was used to determine which have similar proportions of sands but are differentiated by
the major element concentrations after LOI measurement. The their quantities of silts and clays (Table 1). The lower layer
sample powders were first ignited and melted with a lithium tetra- (MD1) is gray with 1.3–1.6 m thick (Table 1). The middle layer
borate flux and then analyzed with a Rigaku RIX-3000 wavelength- (MD2) is dark gray with yellowish brown patches, and is 0.5–
dispersive X-ray Fluorescence spectrometer. Inductively Coupled 1.1 m thick. Both layers have a sandy clay texture (Table 1). The
Plasma-Mass Spectrometry (ICP-MS) was used to quantify trace upper layer (MD3) is 0.5–0.9 m thick, yellowish brown with sandy
and rare-earth element concentrations. Sample powders for heavy clay texture (Fig. 2c), defined by its richness in clays relative
ICP-MS analyses were digested with a mixture of HCl and HClO4 to silts (Table 1).
at 120 °C in sealed Teflon containers for one week, and then rinsed The heavy mineral fractions of the samples are mainly com-
out of their containers with dilute HNO3 and dried. The residue was posed of epidote, olivine, and opaque minerals, and accessory gar-
again dissolved in a mixture of three acids (HNO3, HCl and HF) at net and zircon (Table 2). These minerals have essentially angular
100 °C. Sample solutions were analyzed in a Perkin Elmer Elan forms that may indicate moderately weathered materials and very
9000 ICP-MS instrument. close proximity of the source rocks, such as the granitoids defined

Table 1
Particle size distribution (weight in %) in different layers of the Ngaye River sediments.

Ref. code Coarse sands Fine sands Silts Clays Total Texturea
Upper layer MD3 16.99 39.01 0.83 43.15 99.98 Sandy heavy clay
Middle layer MD2 14.47 43.34 8.24 33.93 99.98 Sandy clay
Lower layer MD1 7.20 48.70 16.50 27.30 99.70 Sandy clay

Coarse sands: 2000 < U < 200 lm; fine sands: 200 < U < 50 lm; silts: 50 < U < 2 lm; clays: U < 2 lm.
a
From Ngon Ngon et al., 2009.

Table 2
Mineral distribution of the heavy mineral concentrates in Ngaye River sediments.

Ref. code Epidote Olivine Opaque minerals Augite Zoisite Diopside Garnet Zircon Tourmaline Non-identified
Upper layer MD3 12.00 29.00 45.00 0.50 – 1.00 4.00 3.00 1.00 4.50
Middle layer MD2 13.00 40.00 20.00 0.50 0.50 – 5.00 1.00 0.50 19.50
Lower layer MD1 32.50 29.50 9.00 4.50 5.00 2.50 2.00 1.00 0.50 12.00

Table 3
Mineralogical composition of Ngaye River sediments.

Lower sandy clay layer – MD1 Middle sandy clay layer – MD2 Upper sandy heavy clay layer – MD3
Samples Bulk Coarse Fine Silts Clays Bulk Coarse Fine Silts Clays Bulk Coarse Fine Silts Clays
sediment sands sands sediment sands sands sediment sands sands
Ref. code MD11 MD14 MD12 MD15 MD13 MD20 MD24 MD23 MD22 MD21 MD30 MD34 MD33 MD32 MD31
Smectite ++ - + ++ ++ – – – + + + – – ++ +++
Kaolinite ++ + + ++ ++ ++ – – ++ +++ ++ e – ++ +++
Illite + + ++ + + +++ ++ ++ +++ ++ + – + ++ ++
Quartz ++ +++ +++ + + +++ +++ +++ + + ++ +++ +++ + +
Feldspars (+ + ++ + – – + + + – – + ++ ++ – –
Plagioclases)
Hematite – – – – – + – – – – – – – – –
Goethite – – – – – – – – – – + – – – –
Ilmenite – – + – – + + + – – – + + – –
Amphibole – – + – – + + ++ – – + + + – –
Olivine + – + – – – – – – – – – – – –
Rutile + + – – – + + + – – + – + – –
Pyroxene – – – – – + + + – – + – + – –
Zircon – – – – – – – – – – – – + – –

+++: very abundant; ++: abundant; +: poorly represented; e: trace; –: not identified.
168 P.-D. Ndjigui et al. / Journal of African Earth Sciences 100 (2014) 164–178

in Fig. 1. Each sample shows different heavy minerals proportions abundant minerals in the finest sediments are kaolinite, illite and
with epidote and olivine dominating in the sandy clay layers, while smectite (Table 3). Fig. 3 confirms the high proportion of kaolinite
opaque minerals increase in proportion from the lower to upper and illite in the oriented sub-samples of clay- and silt-sized frac-
layers (Table 2). tions. Smectite mainly occurs in the fine fractions of the upper
The bulk mineralogical compositions of samples reveal the layer (Fig. 3e-f).
occurrences of smectite, kaolinite, quartz, illite and few amounts
of rutile, amphibole, olivine, pyroxene, feldspars, ilmenite, goethite 4.2. Geochemical characterization
and hematite (Table 3). The sand fractions show the same minerals
as the bulk sediments but with a higher quartz proportion (Table 3). 4.2.1. Main geochemical patterns
Note that amphibole amount may be abundant with some pyrox- The main geochemical patterns of the Ngaye River alluvial sed-
ene and olivine in the sandy clay layers (MD1 and MD2). The most iments are first defined by a principal component analysis (PCA) of

clay-sized fraction silt-sized fraction

illite
illite + kaolinite
I tensity (arbitrary unit)

Intensity (arbitrary unit)

illite + kaolinite

smectite

kaolinite

illite + quartz
illite

kaolinite
smectite

illite + quartz

kaolinite
kaolinite

quartz
quartz
n

a 2.5 12.5 22.5 32.5 b 2.5 12.5 22.5 32.5

Degree 2θ Degree 2θ
illite
illite + kaolinite
Intensity (arbitrary unit)
Intensity (arbitrary unit)

illite + kaolinite

illite + quartz
illite + quartz

kaolinite
illite

smectite
kaolinite
smectite

kaolinite

kaolinite
quartz

quartz

c 2.5 12.5 22.5 32.5 d 2.5 12.5 22.5 32.5

Degree 2θ Degree 2θ
illite + kaolinite
illite + kaolinite

illite
illite

Intensity (arbitrary unit)

Intensity (arbitrary unit)

smectite

illite + quartz
kaolinite
smectite

illite + quartz
kaolinite

kaolinite
kaolinite

quartz
quartz

e 2.5 12.5 22.5 32.5 f 2.5 12.5 22.5 32.5

Degree 2θ Degree 2θ

Untreated Treated with ethylene glycol

Treated by heating (500-600 °C) Treated with hydrazine

Fig. 3. X-ray diffraction analyses of the finest grain-sized fractions after specific treatments (Left column: clay- and right column silt-sized fractions): (a and b) lower sandy
clay layer; (c and d) Middle sandy clay layer; (e and f) upper sandy heavy clay layer.
 P.-D. Ndjigui et al. / Journal of African Earth Sciences 100 (2014) 164–178 169

60

Ti-Zr heavy minerals

Factor 2 (39.3 % of total variance) 40
TiO2, Zr, Y, Nb, Hf
quartz + plagioclases Th, U,Ta, Cd, W
La, Ce, Pr, Nd, Sm, Eu
SiO2, CaO, Na2O Gd, Tb, Dy, Ho, Er, Tm, Yb, Lu
20 Ba, Sr

K2O, Rb Al2O3, Fe2O3, MnO, MgO,

P2O5, H2O
K-felspars Cr, V, Ni, Zn, Cu, Co, Sc
Li, Ga, Pb, Mo, Be, Cs, Tl
-20

clays + Fe-Mn oxihydroxides

-40
-80 -60 -40 -20 0 20 40 60 80 100

Factor 1 (41.9 % of total variance)

chemical elements
sandy clay (MD1) sandy clay (MD2) sandy heavy clay (MD3)

bulk coarse sands fine sands silts clays

Fig. 4. Principal component analysis (PCA) diagram of the geochemical data, each sample being projected in the F1–F2 plan.

the geochemical compositions of the different grain-sized fractions more siliceous than the expected granitoid source rocks (BG, BMG
of each sediment layer (Fig. 4). The first two factors F1 and F2 and HBG) while the finest sediments are more aluminous
describe 41.9% and 39.3% of the total variance, respectively (Fig. 5a). Ti and P are enriched in the silt- and clay-sized fractions,
(Fig. 4). The projection diagram F1–F2 allows the discrimination but Ti content is higher in the fine sand fraction of the upper sandy
of SiO2, CaO, Na2O (quartz, plagioclases) and K2O, Rb (K-feldspars) heavy clay layer (MD3) indicating the presence of rutile and/or
from TiO2, Zr, REE (Ti–Zr heavy minerals) and Al2O3, Fe2O3, MnO, ilmenite (Tables 3 and 4). The silt–clay and sand fractions have
MgO, P2O5, LOI and the other trace elements (clays and Fe–Mn oxi- equivalent (CaO + Na2O + K2O) contents, but differentiated Al2O3
hydroxides). The projection F1–F2 splits the four above mentioned contents (Fig. 5b). A positive trend occurs for the sand fractions
geochemical groups in separate fields (Fig. 4). Based on this PCA and the expected granitoid rocks source, while Al2O3 and
analysis and the projection of the samples in the F1–F2 diagram, (CaO + Na2O + K2O) are clearly negatively correlated for the silt–
the different size fractions of the alluvial sediments are character- clay fractions (Fig. 5b), suggesting that feldspars are little weath-
ized by specific or a mixing of the geochemical signatures (Fig. 4). ered in the finest sediments and well preserved with quartz in
The silt- and clay-sized fractions of the Ngaye River sediments are the sand fractions (Fig. 5c). This moderate weathering of the source
relatively enriched in Al2O3, Fe2O3, MgO, P2O5, MnO and transition rocks agrees with limited depletion of Ca, K and Na relative to Al2O3
elements (V, Cr, Ni, Co, Cu and Zn) including also Sc, Pb, Li, Ga, Cs, and Fe2O3 (Fig. 5b–d). MgO, Fe2O3 and the Chemical Index of
Be and Mo, the silt fraction and the bulk of the lower sandy clay Alteration (CIA) are negatively correlated with the SiO2/Al2O3 ratio
layer (MD1) are also imprinted by K2O and Rb. These fractions (Fig. 5e–g), implying that Fe2O3 content increases with increasing
are also rich in kaolinite and smectite (Fig. 3, Table 3). All the weathering (Fig. 5h). In Fig. 5g, SiO2/Al2O3 comprised between 1.5
coarse sand fractions with the fine sands of the lower layer and 2.5 for CIA > 80% indicates that kaolinite and smectite (e.g.,
(MD1) and the bulk samples of middle and upper layers (MD2 Mg-smectite, see Fig. 5 and Fig. 5g) are the most abundant clay
and MD3) are placed between quartz and plagioclases and minerals in the finest fractions (Table 3, Fig. 3).
K-feldspars. Ti, Zr, Hf, Th, U, Y and REE characterize the fine sand
fractions of the middle and upper layers, and indicate a heavy
mineral signature (Fig. 4). 4.2.3. Trace element distribution patterns
Geochemical composition of trace element normalized to PAAS
(McLennan, 1989) provide an overview of the geochemical pat-
4.2.2. Major chemical element patterns terns of the different grain-sized fractions in each layer (Table 5,
The Ngaye River sediments have moderate to high SiO2 contents Fig. 6a–c). The coarse sand fraction from the lower and middle lay-
(38.06–91.72 wt.%), with a mean of 64.89 wt.% SiO2 (Table 4). ers is strongly depleted in U, Th, La, Ce, Nd, Sm and Yb, and slightly
Al2O3 content increases as SiO2 decreases, the sand fractions being in Rb, Zr, Eu, Ca, Fe, Co and Cr (Fig. 6a). These two layers show
170 P.-D. Ndjigui et al. / Journal of African Earth Sciences 100 (2014) 164–178

Table 4
Major element composition (wt.%) and element ratios of Ngaye River sediments.

Samples d.l. Lower sandy clay layer – MD1 Middle sandy clay layer – MD2 Upper sandy heavy clay layer – MD3
Bulk Coarse Fine Silts Clays Bulk Coarse Fine Silts Clays Bulk Coarse Fine Silts Clays
sediment sands sands sediment sands sands sediment sands sands
Ref. code – MD11 MD14 MD12 MD15 MD13 MD20 MD24 MD23 MD22 MD21 MD30 MD34 MD33 MD32 MD31
SiO2 0.01 55.05 83.27 74.02 48.01 34.71 75.04 91.72 85.45 43.31 38.06 67.96 89.76 81.29 46.12 41.26
Al2O3 0.01 19.80 6.91 12.65 18.79 22.97 10.89 4.61 6.25 21.94 23.49 13.12 4.92 6.17 21.78 25.14
Fe2O3 0.01 7.16 2.35 2.81 11.05 12.77 4.25 0.62 2.28 12.46 12.58 5.56 1.02 4.30 10.47 11.37
MnO 0.01 0.07 0.02 0.03 0.11 0.12 0.04 0.01 0.05 0.12 0.09 0.05 0.01 0.09 0.10 0.09
MgO 0.01 1.21 0.79 1.07 1.90 1.22 0.85 0.10 0.67 1.73 1.20 0.64 0.12 0.50 1.18 1.10
CaO 0.01 1.04 0.72 2.14 1.06 0.56 1.07 0.58 1.13 0.59 0.40 0.83 0.67 1.31 0.72 0.57
Na2O 0.01 0.92 1.14 2.47 0.85 1.22 1.08 0.77 0.77 0.64 0.97 0.73 0.99 1.05 0.46 0.35
K2O 0.01 1.78 1.86 2.36 2.22 1.23 2.00 1.49 1.29 1.97 1.34 1.85 1.47 1.35 1.72 1.08
TiO2 0.01 0.82 0.29 0.44 1.03 0.99 0.48 0.06 0.74 1.18 1.07 0.64 0.21 2.17 1.04 0.99
P2O5 0.01 0.07 0.08 0.05 0.58 3.32 0.04 0.02 0.06 1.17 2.63 0.06 0.03 0.19 0.53 0.79
LOI 0.05 11.16 2.46 1.92 13.51 19.50 3.87 0.34 0.39 14.35 17.62 7.50 0.34 0.50 14.70 16.45
Total – 99.08 99.89 99.96 99.11 98.61 99.62 100.31 98.98 99.46 98.94 98.94 99.55 98.91 98.82 99.19
CIA (%) – 84.11 65.00 64.48 88.41 81.98 72.41 61.88 66.21 87.27 89.66 79.37 61.12 62.45 88.25 92.63
SiO2/Al2O3 – 2.78 12.05 5.85 2.55 1.51 6.89 19.89 13.67 1.97 1.62 5.17 18.24 13.17 2.12 1.64

d.l.: detection limits;

LOI: Loss of ignition.
CIA (%) = (Al2O3/(Al2O3 + CaO + Na2O + K2O))  100 from Nesbitt and Young (1984).

comparable trends but with a slight enrichment of Th, La, Ce, Nd, Zr concentrated in the silts and clays (Fig. 7c–d, Table 5). These frac-
and Sm and a large K negative anomaly in the fine sand fraction of tions are characterized by ratios Th/Y < 1, Zr/Y < 10 (Fig. 7e) and,
MD2 (Fig. 6b). The upper sandy heavy clay layer (MD3) shows the Sr/Ni 6 1 and Y/Hf P 10 (Fig. 7f).
same trend except the coarse fraction that is only depleted in Yb,
Fe, Co, Cr (Fig. 6c). The Ngaye River sediments geochemical compo- 4.2.4. Rare-earth element fractionation
sition normalized to chondrite (McDonough and Sun, 1995) reveal Total REE content varies between 29 and 1851 ppm in the fine
that these sediments are strongly enriched in many elements sands, with moderate concentrations (180–282 ppm) in the bulk
except Fe, Co and Cr, and characterized by negative anomalies in sediments and silt–clay fractions (Table 6). La, Ce and Nd have
K and Na (Fig. 6d–f). These anomalies and their chemical trends higher contents than other REE, particularly in the fine sands.
are concordant to the spider diagram forms of the expected gran- The total content of these three lanthanides vary between 21 and
itoid rock source (Fig. 6g). 1609 ppm, with less than 500 ppm in several samples (Table 6).
Lithophile elements Y, Nb, Hf, Th, U, Ta, Cd and W are associated Amongst HREE, only Gd and Dy have relatively high concentrations
to TiO2 and Zr, i.e., to ilmenite/rutile in the residual sand fractions (>3 ppm) in the different size fractions. REE currently distribute
(Fig. 4). Y, Hf, Nb, and Ta contents are very low and Zr content var- between Ti and Zr heavy minerals (Fig. 4). HREE are rather carried
ies between 14 and 5341 ppm (Table 5). However, the highest con- by ilmenite, rutile and monazite. LREE are mostly carried by zircon
tent of several trace elements are observed in the fine sand fraction and as previously mentioned Zr is highly concentrated in the fine
of the upper layer (Table 5) in relation with heavy mineral sand fractions (Table 5).
concentrations. REE concentrations normalized to PAAS (McLennan, 1989) indi-
Transition elements with Li, Ga, Pb, Mo, Be, Cs and Tl are cate: (i) REE-enrichment except in the coarse sands; (ii) slight
rather associated to Al2O, Fe2O3, MnO, MgO and P2O5 that charac- LREE-enrichment and HREE-depletion; (iii) positive Eu anomalies
terized the silt and clay fractions (Fig. 4, Tables 4 and 5). Sr and (Eu/Eu 1.34 to 3.96), particularly for the coarse sand fractions
Ba exhibit similar behavior and are linked to SiO2, CaO and of the sandy clay layers (Fig. 8a and b); (iv) large negative Eu
Na2O, i.e., to the bulk sediment and to residual plagioclases anomalies (Eu/Eu 0.35 to 0.70) for the fine sand fractions of
concentrated in the sand fractions (Fig. 4, Tables 4 and 5). The the middle and upper layers (MD2 and MD3) and the coarse sand
concentration of transition trace elements such as Ni and Cr is fraction of the upper layer (Fig. 8b and c); (v) REE slight fraction-
low relative to Ba and Sr contents, although Cr, V, Ni, Zn, Co ation between the different grain-size (LaN/YbN varying between
and Li have higher concentrations in the clay fractions (Fig. 7a, 0.72 and 7.37); and (vi) low (La/Sm)N and (Gd/Yb)N ratios (Table 6).
Table 5). Ba content varies from 230 to 1499 ppm, with highest Several samples have similar PAAS-normalized REE patterns with
concentration in the coarse sand fraction of the lower sandy clay high abundances (Fig. 8a–c). The chondrite-normalized REE
layer (Table 5). Ba/Cr and Sr/Ni decrease as CIA increases suggest- patterns confirm the similar high LREE-enrichment and Eu anom-
ing the preferential retention of Ni and Cr in the finest grain-sized alies for studied samples and expected granitoid source-rocks
fractions (Fig. 7b and c). (Fig. 8d–g). The fine sand fractions of the middle and upper layers
Zr, Ti and Th are relatively enriched in weathering profiles (MD2 and MD3) have REE patterns comparable to those of BG and
(Beauvais and Colin, 1993; Braun et al., 1993; Ndjigui et al., BMG (Fig. 8e–g), while the REE pattern of the finest fractions (silts
2008, 2009) and also in sediments by heavy minerals sorting pro- and clays) rather underlines the signature of hornblende-biotite
cess. As such Zr and Th are used for testing the behavior of some granitoids, HBG (Fig. 8d–g).
trace elements. Regardless of its content in the expected source HREE are concentrated in the finest fractions (silts and clays)
rocks, Zr increases in the sand fractions and decreases in the finest rather than in the sand fractions (Fig. 9a–f, Table 6). The finest frac-
sediments (see Fig. 6a–c) with increasing Y content and CIA tions are slightly depleted in LREE but relatively enriched in HREE
(Fig. 7d). Zr/Y increases with Th/Y (Fig. 7e), indicating that Th is that may be correlated to increasing Ti content and CIA in the same
preferentially hosted by zircon in the fine sand fractions (Table 5). fractions (Fig. 9a and b). Conversely, the sand fractions are enriched
Ni and Y increase with CIA relatively to Sr and Zr and are mainly in LREE relative to HREE (LREE/HREE 6.59 to 21.63). The highest
 P.-D. Ndjigui et al. / Journal of African Earth Sciences 100 (2014) 164–178 171

90 25

80 20

Al2O3, wt. %
SiO2, wt. %
70
15
60
10
50

40 5

CaO+Na2O+K2O, wt. %
a 30 b 0
0 5 10 15 20 25 30 0 2 4 6 8 10 12 14
Al2O3, wt. %
15 15
Al2O3 / (CaO+Na2O+K2O)

Al2O3 / (CaO+Na2O+K2O)
10 10

5 5

Fe2O3, wt. %
c 0 d 00
30 40 50 60 70 80 90 100 5 10 15
SiO2, wt. %

20 20

15 15
SiO2/Al2O3
SiO2/Al2O3

10 10

5 5

0
f 0
e 0 0.5 1 1.5 2 2.5 3 3.5 0 5 10 15
MgO, wt. % Fe2O3, wt. %
15
20
Fe2O3, wt. %
SiO2 / Al2O 3

15 10

10
5
5

CIA %
g 0 0
55 60 65 70 75 80 85 90 95
h 55 60 65 70 75 80 85 90 95
CIA %
sandy clay (MD1) sandy clay (MD2) sandy heavy clay (MD3) BMG BG HBG

bulk sediments coarse sands fine sands silts clays

Fig. 5. Geochemical Harker diagrams of selected major elements: (a) SiO2 wt.% vs. Al2O3 wt.%; (b) Al2O3 wt.% vs. (CaO + Na2O + K2O) wt.%; (c) Al2O3/(CaO + Na2O + K2O) vs.
SiO2 wt.%; (d) Al2O3/(CaO + Na2O + K2O) vs. Fe2O3 wt.%; (e) SiO2/Al2O3 vs. MgO wt.%; (f) SiO2/Al2O3 vs. Fe2O3 wt.%; (g) SiO2/Al2O3 vs. CIA%; (h) SiO2/Al2O3 vs. CIA% (HBG:
hornblende-biotite granitoids; BG: biotite granitoids and BMG: biotite ± muscovite granitoids).
172 P.-D. Ndjigui et al. / Journal of African Earth Sciences 100 (2014) 164–178

10 4

10 10 3

Sample / chondrite
sandy clay (MD1) sandy clay (MD1)
Sample / PAAS

10 2

1 10

10 -1 10 -1

10 -2

10 -2 10 -3
a Rb U Th K La Ce Nd Zr Sr Na Sm Eu Yb Ca Fe Co Cr d Rb U Th K La Ce Nd Zr Sr Na Sm Eu Yb Ca Fe Co Cr

10 4
10 sandy clay (MD2) sandy clay (MD2)

Sample / chondrite
10 3
Sample / PAAS

10 2
1
10
1
10 -1
10 -1

10 -2
10 -2 10 -3
b Rb U Th K La Ce Nd Zr Sr Na Sm Eu Yb Ca Fe Co Cr e Rb U Th K La Ce Nd Zr Sr Na Sm Eu Yb Ca Fe Co Cr

10 4 sandy heavy clay (MD3)

Sample / chondrite

sandy heavy clay (MD3)

10 3
Sample / PAAS

10 2
10
10
1 1

10 -1
10 -1
10 -2

10 -2 10 -3
c Rb U Th K La Ce Nd Zr Sr Na Sm Eu Yb Ca Fe Co Cr f Rb U Th K La Ce Nd Zr Sr Na Sm Eu Yb Ca Fe Co Cr

10 4
Granitoïds
Sample / chondrite

bulk sediments 10 3
coarse sands 10 2
fine sands
10
silts
1
clays
Biotite-muscovite granitoids (BMG) 10 -1

Biotite granitoids (BG) 10 -2

Hornblende-biotite granitoids (HBG) 10 -3

g Rb U Th K La Ce Nd Zr Sr Na Sm Eu Yb Ca Fe Co Cr

Fig. 6. Geochemical spider diagrams of the Ngaye River sediments: (a–c) PAAS-normalized (McLennan, 1989); (d–g) Chondrite-normalized (McDonough and Sun, 1995).

LREE-enrichment in the fine sands is linked to the highest Zr content sized fractions that suggest REE fractionation with increasing
in the expected host granitoids, e.g., BMG (Fig. 9c). The highest weathering (Fig. 9d). The coarse sand fractions have higher Eu/Eu
HREE-enrichment in silt and clay fractions is linked to lowest Zr (3–4) than the finest sediments (Eu/Eu <1.5), which are character-
and highest Y contents characterizing the greenstones (BG and ized by GdN/YbN <2 (Fig. 9e). Granitoid signature is characterized
HBG). LaN/YbN varies from 18 to 80 in granitoids, and slightly by Eu/Eu <1.5, GdN/YbN P2, and CeN/YbN >1.5, which is potentially
fractionate around LaN/YbN 2 between the coarse and fine grain- recorded by the fine sand fractions (Fig. 9e and f).
 P.-D. Ndjigui et al. / Journal of African Earth Sciences 100 (2014) 164–178 173

102
10 2

Ba/Cr

Ba/Cr
10 10

Sr/Ni
1 1
a 10-1 1 10 102 b 55 60 65 70 75 80 85 90 95
CIA %
10 2 2
10

10
Sr/Ni

Zr/Y
10

10 -1
c 60 65 70 75 80 85 90 95 d 55 60 65 70 75 80 85 90 95
CIA % CIA %

10 102

10
Sr/Ni
Th/Y

Zr/Y
10-1 10-1
e 1 10 102 f 1 10 102
Y/Hf

sandy clay (MD1) sandy clay (MD2) sandy heavy clay (MD3) BMG BG HBG

bulk sediments coarse sands fine sands silts clays

Fig. 7. Geochemical scattergrams of (a) Ba/Cr vs. Sr/Ni; (b) Ba/Cr vs. CIA%; (c) Sr/Ni vs. CIA%; (d) Zr/Y vs. CIA%; (e) Th/Y vs. Zr/Y; (f) Sr/Ni vs. Y/Hf (see Fig. 5 for HBG, BG and
BMG acronyms).

5. Discussion Fig. 7 further suggest that these granitoid rocks are the main
sources of the Ngaye River alluvial sediments. However, some tran-
Variations of the mineralogical, geochemical composition and sition elements and REE patterns may also suggest to a certain
sorting of the alluvial sediments of the Ngaye River document extent a potential origin of the sediments from greenstones and/
the source rock types including their weathering degree. The distri- or basic volcanics (e.g., Figs. 6d–g, 8d–g and 9e–f).
bution of chemical elements in the bulk sediment and the different
grain-sized fractions and the relatively low CIA values suggest sed- 5.1. Source rock geochemical signature of alluvial sediments
iment sources from the lower less altered parts (arenaceous sapr-
olites) of weathering profiles developed mainly upon granitoid Chemical element distribution in alluvial sediments depends
rocks (e.g., Fig. 5a–d). The geochemical patterns displayed in on their grain size and mineralogical composition (Shimizu and
174 P.-D. Ndjigui et al. / Journal of African Earth Sciences 100 (2014) 164–178

Table 5
Trace element composition (ppm) and element ratios of Ngaye River sediments.

Samples d.l. Lower sandy clay layer – MD1 Middle sandy clay layer – MD2 Upper sandy heavy clay layer – MD3
Bulk Coarse Fine Silts Clays Bulk Coarse Fine Silts Clays Bulk Coarse Fine Silts Clays
sediment sands sands sediment sands sands sediment sands sands
Ref. – MD11 MD14 MD12 MD15 MD13 MD20 MD24 MD23 MD22 MD21 MD30 MD34 MD33 MD32 MD31
code
Cr 3.00 135.00 45.00 72.00 101.00 160.00 68.00 15.00 47.00 165.00 157.00 80.00 36.00 93.00 152.00 164.00
V 0.80 116.00 30.50 45.40 95.70 95.60 68.70 11.00 13.20 184.10 188.30 86.60 15.80 71.10 167.50 169.70
Ni 1.60 73.20 29.20 31.10 55.10 56.00 27.70 5.10 12.00 83.90 89.70 37.60 7.30 15.60 78.40 89.60
Zn 7.00 72.00 35.00 38.00 63.00 60.00 40.00 8.00 20.00 109.00 108.00 43.00 16.00 36.00 88.00 95.00
Cu 1.40 49.50 26.00 23.30 38.50 118.40 20.60 7.30 10.10 96.10 101.30 29.10 14.20 16.30 72.10 69.60
Co 0.13 24.85 12.81 11.84 19.00 17.54 14.31 1.95 5.03 42.36 32.13 16.43 2.21 6.49 33.53 35.07
Sc 1.10 14.50 3.40 6.30 13.20 14.20 8.30 2.40 7.80 14.10 20.30 10.10 2.40 6.90 21.20 19.00
Ba 0.80 947.90 982.10 1499.20 397.50 314.90 899.30 792.90 674.90 380.50 365.70 853.90 912.90 846.90 712.70 230.40
Zr 6.00 149.00 85.00 299.00 56.00 14.00 321.00 135.00 901.00 79.00 56.00 330.00 246.00 5341.00 83.00 44.00
Sr 0.60 147.20 152.00 318.10 47.20 30.50 144.60 118.80 112.70 41.70 27.20 133.30 155.60 174.10 91.40 37.60
Y 0.05 17.77 2.18 8.65 15.56 16.09 11.70 2.13 24.73 21.70 36.90 17.19 5.94 53.49 26.46 15.92
Li 0.40 15.90 6.10 5.30 18.20 19.30 5.30 1.70 1.70 14.80 21.90 8.40 1.80 1.60 22.90 25.80
Ga 0.04 26.01 7.96 14.36 19.77 18.71 12.95 4.21 7.27 30.02 31.38 16.05 4.56 8.07 29.22 32.39
Pb 0.60 17.20 9.60 14.20 11.90 12.00 13.20 9.40 11.10 20.30 20.20 15.00 9.00 17.80 20.00 17.10
Rb 0.23 41.73 43.37 49.54 53.72 44.19 44.83 27.40 23.02 11.50 34.57 36.90 25.73 22.22 72.22 17.83
Nb 0.03 11.71 3.27 5.56 7.99 3.01 10.26 1.43 16.50 18.93 17.58 10.02 2.32 31.87 16.96 16.03
Hf 0.14 3.84 2.14 7.36 1.08 0.29 8.50 3.60 24.65 2.20 1.53 9.15 5.95 >29 2.27 1.25
Mo 0.08 0.78 0.33 0.26 15.85 19.55 0.69 0.14 0.24 3.05 4.06 0.85 0.12 0.45 1.87 2.19
Th 0.02 14.51 1.65 14.31 9.47 10.45 17.12 2.39 62.83 11.93 18.68 13.48 19.99 224.00 15.39 12.68
U 0.01 1.21 0.21 1.17 0.85 0.87 1.41 0.34 4.28 1.49 1.31 1.47 1.20 15.20 1.40 1.10
Ta 0.02 0.61 0.18 0.35 0.34 0.11 0.84 0.11 1.71 1.15 0.95 0.61 0.17 2.81 0.96 0.90
Be 0.04 1.72 0.44 0.73 1.32 1.35 0.60 0.29 0.41 1.38 1.73 0.90 0.33 0.39 1.69 1.85
Sn 0.16 1.52 0.54 0.54 1.45 1.26 0.84 0.21 0.68 2.13 2.33 1.00 0.22 0.84 1.81 2.02
W 0.05 0.38 0.31 0.31 0.28 0.22 0.40 0.08 0.78 0.53 0.52 0.40 0.09 1.38 0.54 0.51
Tl 0.01 0.33 0.19 0.21 0.28 0.23 0.26 0.11 0.10 0.56 0.48 0.29 0.10 0.09 0.48 0.45

d.l: detection limits.

Masuda, 1977; Sawyer, 1986; Kasanzu et al., 2008; Bhuiyan 5.2. Sorting and weathering of alluvial sediments
et al., 2011). As such, quartz-rich alluvial sediments may result
from the erosion of large tracts of granitoids in the Ngaye River Sorting of alluvial sediment determines the mineral distribution
watershed. The high Ba and Sr contents (Fig. 7a and b) are indic- in the different grain-sized fractions that also primarily depends
ative of feldspars (including plagioclases) depending on the gran- upon the composition and weathering degree of source rocks.
itoid nature of source rocks (Tchameni et al., 2006). Mg, Cr and Unaltered feldspars and pyroxene attest of the sediment immatu-
Sc contents are rather indicative of ferromagnesian minerals rity due to short distance transport from moderately weathered
such as biotite, pyroxene, amphibole and olivine (Nesbitt, source rocks (Hofmann, 2005; Das et al., 2006; Marques et al.,
1979), and chromite, which are current minerals of greenstones 2011). Illite is inherited from such source rocks as well as kaolinite
(see Fig. 1b). High proportion of quartz in the alluvial sediments and smectite. These clay minerals may have also neo-formed by
may explain the low trace element contents (including REE) in weathering process of primary minerals in the alluvial sediments.
the coarse sand fractions (Das et al., 2006; Tchameni et al., For example, smectite may derive from the chemical weathering
2006). of residual Fe–Mg minerals.
The overall REE distribution in the Ngaye River sediments Weathering and sorting of minerals, such as feldspars and/or
reflects the REE distribution in the granitoid rocks (Fig. 8a–g). amphibole and pyroxene control Ba and Sr content variations
The moderate REE contents in the clay-sized fractions depend on (Das et al., 2006; Rahman and Suzuki, 2007; Harlavan et al.,
clay mineral species. The LREE-enrichment may be linked to Zr 2009; Bhuiyan et al., 2011), relatively to the transition elements,
and/or epidote (Gromet and Silver, 1983), which are concentrated e.g., Ni and Cr, in the different grain-sized fractions (Fig. 7a–c).
in the sand fractions of the upper sandy heavy clay layer (Fig. 8b, Chemical weathering process in sediments currently removes
Tables 2 and 5). Large Eu positive anomaly (Eu/Eu 3.2 to 4) labile cations, e.g., Ca2+, Na+, K+, and concentrates relatively stable
of the coarse sand fractions of the sandy clay layers (Figs. 8d–e residual constituents (Fedo et al., 1995), e.g., Fe, Al, Ti. Limited
and 9e), might be potentially a signature of the andesitic basalts depletion of Na, Ca and K are linked to the moderate weathering
located in the southwest part of the Ngaye River watershed of feldspars (including plagioclases), while high Al2O3 content indi-
(Fig. 1a). Large positive Eu anomaly was also described in mafic cates high proportion of micas and clays (Das et al., 2006; Marques
Tertiary volcanics from Northwest Cameroon (Temdjim et al., et al., 2011). Highest concentrations of CaO in the sand fractions
2004; Kamgang et al., 2013). REE distribution patterns of Pan- (e.g., fine sands) may also be attributed to accessory mineral con-
African granitoids however show a negative Eu anomaly (Fig. 8g) centrates (e.g., epidote, see Table 2). Moderate TiO2 and Fe2O3 con-
suggesting that these rocks are potentially the main sources of tents in sands could be related to ilmenite (Zhou et al., 2004).
the Ngaye River sediments. The Eu anomalies might also reflect dif- Decreasing SiO2/Al2O3 ratio, MgO content and increasing Fe2O3
ferential enrichment by mineral sorting of resistant heavy minerals content (Fig. 5e–h) may also reflect the differentially weathered
(e.g., ilmenite, rutile) and/or moderately weathered feldspars residual Fe–Mg minerals such as olivine and/or amphibole and
(Leybourne et al., 2006; Lee et al., 2009) or differences in weather- pyroxene, which are inherited from moderately weathered
ing degree of the inherited primary mineral assemblage (e.g., granitoids. High Fe2O3 content may also reflect the formation of
feldspars, plagioclases and Fe–Mg minerals) in the sedimentary Fe–Mg clays such as smectite in the finest sediment grain-sized
layers. fractions and in a lesser degree Fe-oxihydroxides.
 P.-D. Ndjigui et al. / Journal of African Earth Sciences 100 (2014) 164–178 175

103

Sample / chondrite
10 sandy clay (MD1) sandy clay (MD1)
Sample / PAAS

102
1

10
10 -1

10 -2 1
a d
La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

103
10
sandy clay (MD2) sandy clay (MD2)

Sample / chondrite
Sample / PAAS

1 102

10 -1 10

10 -2 1
b e
La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

104
10 2
Sample / chondrite

sandy heavy clay (MD3) sandy heavy clay (MD3)

103
Sample / PAAS

10
102

1
10

10 -1 1
c f
La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

103
Granitoïds
bulk sediments
Sample / chondrite

coarse sands
fine sands 102
silts
clays
Biotite-muscovite granitoids (BMG) 10
Biotite granitoids (BG)
Hornblende-biotite granitoids (HBG)
1
g
La Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

Fig. 8. REE patterns of the Ngaye River sediments: (a–c) PAAS-normalized (McLennan, 1989); (d–g) Chondrite-normalized (McDonough and Sun, 1995).

6. Conclusion greenstone signature but a contribution from basalts may not be

totally dismissed. Three sedimentary layers were differentiated
The Ngaye River alluvial sediments result from the erosion of by their granulometric texture: sandy clay and sandy heavy clay
moderately weathered proximal granitoids with a marked layers. The mineral assemblages characterizing the different
176 P.-D. Ndjigui et al. / Journal of African Earth Sciences 100 (2014) 164–178

35 35

30 30

LREE / HREE
25

LREE / HREE
25
20
20
15
15
10

5 10

0
Ti/Zr 5
a b 65 70 75 80 90 95
1 10 102 55 60 85
CIA %

35 102

30
LREE / HREE

LaN / YbN
25
10
20

15

10 1

Zr/Y
c 5
1 10 102 d 55 60 65 70 75 80 85 90 95
CIA %
8
4 7

6
CeN / YbN

3
Eu/Eu*

5
4
2
3

2
1
1
GdN / YbN
0 0
e 0 1 2 3 4 5 6 f 0 1 2 3 4 5 6
GdN / YbN
sandy clay (MD1) sandy clay (MD2) sandy heavy clay (MD3) BMG BG HBG

bulk sediments coarse sands fine sands silts clays

Fig. 9. Geochemical scattergrams of (a) LREE/HREE vs. Ti/Zr; (b) LREE/HREE vs. CIA%; (c) LREE/HREE vs. Zr/Y; (d) LaN/YbN vs. CIA%; (e) Eu/Eu vs. GdN/YbN; (f) CeN/YbN vs. GdN/
YbN (see Fig. 5 for HBG, BG and BMG acronyms).

grain-sized fractions of these layers are quartz/feldspars (+ heavy fractions, which are the richest in quartz, moderately altered feld-
minerals) in the sand fractions, and kaolinite/smectite (+ illite) in spars and heavy minerals. Chemical weathering fractionates REE
the finest sediments. This mineralogical sorting controls the geo- between the different grain-sized fractions, with highest REE and
chemical fractionation of trace and rare-earth elements, with an LREE contents in the less weathered finest sand fractions, while
obvious dependence on the mineral assemblages including the lowest lanthanide and highest HREE concentrations are measured
weathering degree of minerals in the different grain-sized in the most weathered and finest grain-sized fractions.
fractions.
Transition elements (e.g., V, Cr, Ni, Co, Cu and Sc) are relatively Acknowledgements
enriched in the most weathered silt- and clay-sized fractions,
which are the richest in clay minerals (kaolinite, illite and smec- The authors gratefully acknowledge the experience and field
tite). Lithophile elements (e.g., Na, Ca, Ba, Sr and to a lesser extent assistance of Milan Tchouatcha and Elisé Sababa of the University
K) and Zr are rather concentrated in the less weathered sand of Yaoundé 1, Cameroon. Our fieldwork was facilitated by the
 P.-D. Ndjigui et al. / Journal of African Earth Sciences 100 (2014) 164–178 177

Table 6
Rare-earth element composition (ppm) and element ratios of Ngaye River sediments.

Samples d.l. Lower sandy clay layer – MD1 Middle sandy clay layer – MD2 Upper sandy heavy clay layer – MD3
Bulk Coarse Fine Silts Clays Bulk Coarse Fine Silts Clays Bulk Coarse Fine Silts Clays
sediment sands sands sediment sands sands sediment sands sands
Ref. code – MD11 MD14 MD12 MD15 MD13 MD20 MD24 MD23 MD22 MD21 MD30 MD34 MD33 MD32 MD31
La 0.040 48.35 6.98 38.26 43.88 47.83 42.81 5.92 131.26 25.96 63.15 40.44 41.39 442.19 53.75 15.60
Ce 0.120 99.70 13.18 69.49 81.12 98.27 84.44 10.54 250.79 59.54 126.42 81.45 78.93 850.97 107.64 31.44
Pr 0.010 10.16 1.44 7.97 8.85 9.55 9.35 1.23 27.88 6.67 12.85 9.17 8.80 89.05 11.38 3.37
Nd 0.060 36.61 5.30 28.18 31.09 33.76 33.42 4.30 99.09 25.31 45.53 33.62 30.70 316.09 40.26 12.39
Sm 0.010 6.07 0.79 4.47 5.06 5.27 5.68 0.77 16.21 4.80 8.14 5.62 4.83 50.35 6.78 2.67
Eu 0.003 1.49 0.45 0.95 1.27 1.29 1.04 0.46 1.30 1.25 2.16 1.27 0.57 2.98 1.74 0.80
Gd 0.010 4.50 0.57 3.18 3.93 3.98 4.11 0.58 11.11 4.12 6.71 4.13 3.06 32.57 5.26 2.55
Tb 0.002 0.62 0.07 0.37 0.53 0.53 0.50 0.08 1.30 0.65 1.05 0.58 0.34 3.52 0.77 0.43
Dy 0.009 3.55 0.39 1.85 2.93 3.07 2.54 0.43 6.00 4.27 6.49 3.36 1.50 14.66 4.77 2.77
Ho 0.040 0.67 0.08 0.33 0.56 0.60 0.46 0.08 0.97 0.89 1.30 0.66 0.24 2.15 0.95 0.56
Er 0.007 1.90 0.22 0.90 1.52 1.65 1.24 0.23 2.37 2.63 3.70 1.84 0.57 4.86 2.75 1.68
Tm 0.001 0.27 0.03 0.14 0.21 0.24 0.18 0.04 0.31 0.40 0.53 0.27 0.07 0.66 0.40 0.24
Yb 0.009 1.72 0.25 0.92 1.34 1.47 1.15 0.25 1.97 2.60 3.28 1.76 0.46 4.43 2.52 1.59
Lu 0.002 0.26 0.04 0.14 0.20 0.21 0.18 0.04 0.31 0.39 0.47 0.27 0.07 0.78 0.38 0.24
REE – 215.87 29.79 157.15 182.15 207.73 187.10 24.93 550.87 139.48 281.78 184.44 171.53 1851.26 239.35 76.33
LREE – 202.58 28.14 149.33 171.27 195.97 176.74 23.22 526.53 123.53 258.25 171.57 165.22 1751.63 221.55 66.27
HREE – 13.29 1.65 7.83 11.24 11.76 10.36 1.71 24.34 15.95 23.53 12.87 6.31 63.63 17.80 10.06
LREE/HREE – 15.24 17.06 19.07 15.24 16.66 17.06 13.58 21.63 7.74 10.98 13.33 26.18 27.53 12.66 6.59
La + Ce + Nd – 184.66 25.46 135.93 156.09 179.86 160.67 20.76 481.14 110.81 235.10 155.51 151.02 1609.25 201.65 59.43
a – 0.86 0.86 0.87 0.86 0.87 0.86 0.83 0.87 0.79 0.83 0.84 0.88 0.87 0.84 0.78
Ce/Ce – 1.04 0.96 0.92 0.95 1.06 0.94 0.90 0.96 1.04 1.02 0.98 0.95 0.99 1.00 1.00
Eu/Eu – 1.34 3.16 1.19 1.34 1.33 1.01 3.96 0.46 1.32 1.38 1.24 0.70 0.35 0.88 1.44
(La/Yb)N – 2.08 2.08 3.07 2.41 2.40 2.75 1.74 4.92 0.74 1.42 1.70 6.65 7.37 1.57 0.72
(La/Sm)N – 1.16 1.28 1.24 1.26 1.32 1.10 1.12 1.18 0.79 1.13 1.05 1.25 1.28 1.15 0.85
(Gd/Yb)N – 1.59 1.39 2.09 1.77 1.64 2.16 1.39 3.41 0.96 1.24 1.42 4.03 4.45 0.72 0.97

d.l. detection limits.

a = (La + Ce + Nd)/REE.
Ce/Ce = (Cesample/CePAAS)/(Lasample/LaPAAS)1/2 (Prsample/PrPAAS)1/2.
Eu/Eu = (Eusample/EuPAAS)/(Smsample/SmPAAS)1/2 (Gdsample/GdPAAS)1/2.
(La/Yb)N = (Lasample/LaPAAS)/(Ybsample/YbPAAS).
(La/Sm)N = (Lasample/LaPAAS)/(Smsample/SmPAAS).
(Gd/Yb)N = (Gdsample/GdPAAS)/(Ybsample/YbPAAS).

support of the Cameroon Oil Transportation Company (COTCO). Fedo, C.M., Nesbitt, H.W., Young, G.M., 1995. Unraveling the effects of potassium
metasomatism in sedimentary rocks and palaeosols, with implications for
We also thank Daniel Borschneck for X-Ray Diffraction investiga-
palaeoweathering conditions and provenance. J. Geol. 84, 17–36.
tions and Philippe Dussouillez for GIS support at Cerege (CNRS, Gromet, P.L., Silver, L.T., 1983. Rare earth element distributions among minerals in a
OSU Pytheas), France. At the last, the authors thank the Editor- granodiorite and their petrogenetic implications. Geochim. Cosmochim. Acta 47
in-Chief (Pat Eriksson) and one reviewer (Pr. H.A. Wanas) for (5), 925–939.
Guiraudie, C., 1955. Carte géologique du Cameroun à l’échelle de 1/500 000, feuille
constructive comments that improved our manuscript. de Ngaoundéré Ouest + notice explicative, 23 pp.
Harlavan, Y., Erel, Y., Blum, J.D., 2009. The coupled release of REE and Pb to the soil
labile pool with time by weathering of accessory phases, Wind River Mountains,
References WY. Geochim. Cosmochim. Acta 73, 320–326.
Hofmann, A., 2005. The geochemistry of sedimentary rocks from the Fig Tree Group,
Beauvais, A., 2009. Ferricrete biochemical degradation on the rainforest – savannas Barberton greenstone belt: implications for tectonic, hydrothermal and surface
boundary of Central African Republic. Geoderma 150 (3–4), 379–388. processes during mid-Archaean times. Precambr. Res. 143, 23–49.
Beauvais, A., Colin, F., 1993. Formation and transformation processes of iron Kamgang, P., Chazot, G., Njonfang, E., Tchuimegnie, N.B., Tchoua, F.M., 2013. Mantle
duricrust systems in tropical humid environment. Chem. Geol. 106, 77–101. sources and magma evolution beneath the Cameroon Volcanic Line:
Bhuiyan, M.A., Rahman, J.J., Dampare, S.B., Suzuki, S., 2011. Provenance, tectonics geochemistry of mafic rocks from the Bamenda Mountains (NW Cameroon).
and source weathering of modern fluvial sediments of the Baahmaputra- Gondwana Res. 24, 727–741.
Jamuna River, Bangladesh: inference from geochemistry. J. Geochem. Explor. Kasanzu, C., Matboko, M.A.H., Manya, S., 2008. Geochemistry of fine-grained
111, 113–137. sedimentary rocks of the Neoproterozoic Ikorongo Group, NE Tanzania:
Boulangé, B., Colin, F., 1994. Rare-earth element mobility during conversion of Implications for provenance and source rock weathering. Precambr. Res. 164,
nepheline syenite into lateritic bauxite at Passo-Quatro, Minas Gerais, Brazil. 201–213.
Appl. Geochem. 96, 701–711. Koch, P., 1958. Etude géologique du Cameroun, feuille de Garoua-Ouest + notice
Brabant, P., Gavaud, M., 1985. Les sols et les ressources du Nord-Cameroun. Cartes + explicative, 48 pp.
notice explicative. ORSTOM, 285 pp. Lasserre, M., 1958. Etude géologique de la partie orientale de l’Adamaoua
Braun, J.-J., Pagel, M., Herbillon, A., Rosin, C., 1993. Moblization and redistribution of (Cameroun central) et les principales sources minéralisées de l’Adamaoua.
REES and thorium in a syenitic lateritic profile – a mass balance study. Geochim. Thèse, Univ., Clermont-Ferrand.
Cosmochim. Acta 57 (18), 4419–4434. Lawal, A.O., Abdullahi, Y., 2010. Evaluation of industrial potentials of alluvial clays
Braun, J.-J., Pagel, M., Muller, J.-P., Bilong, P., 1990. Cerium anomalies in lateritic from the confluence of rivers Niger and Mimi. Sci. World J. 5 (3), 17–20.
profiles. Geochim. Cosmochim. Acta 54, 781–795. Le Maréchal, A., 1976. Géologie et géochimie des sources thermominérales du
Das, B.K., Al-Mikihlafi, A.S., Kaur, P., 2006. Geochemistry of Mansar lake sediments, Cameroun. Travaux et Documents. ORSTOM 59, Paris, 176 pp.
Jammu, India: implication for source-area weathering, provenance, and tectonic Le Maréchal, A., Vincent, P.M., 1971. Le fossé crétacé du Sud-Adamaoua (Cameroun).
setting. J. Asian Earth Sci. 26, 640–668. Cah. ORSTOM, sér. Géol., III (1), 67–83.
Dirasset, J., Cornachia, M., Dars, R., 2000. Un trait structural majeur du continent Lee, R.G., Dilles, J.H., Mazdab, F.K., Wooden, J.L., 2009. Europium Anomalies in Zircon
Africain. Les linéaments Centrafricains, du Cameroun au Golfe d’Aden. Bull. from Granodiorite Porphyry Intrusions at the El Salvador Porphyry Copper
Société Géologique France 7, 101–109. Deposit, Chile. The Geological Society of America, Paper n°158–8.
Etame, J., Gerard, M., Suh, C.E., Bilong, P., 2009. Halloysite neoformation during the Letouzey, R., 1985. Notice explicative de la carte phytogéographique du Cameroun à
weathering nephenilitic rocks under humid tropical conditions at Mount l’échelle de 1/500 000. Institut de la Carte Internationale de Végétation,
Etinde, Cameroon. Geoderma 154, 59–68. Toulouse, pp. 240.
178 P.-D. Ndjigui et al. / Journal of African Earth Sciences 100 (2014) 164–178

Leybourne, M.I., Peter, J.M., Layton-Matthews, D., Volesky, J., Boyle, D.R., 2006. Ngon Ngon, G.F., Yongue-Fouateu, R., Bitom, D.L., Bilong, P., 2009. A geological study
Mobility and fractionation of rare earth elements during supergene weathering of clayey laterite and clayey hydromophic material of the region of Yaoundé
and gossan formation and chemical modification of massive sulphide gossan. (Cameroon): a prerequisite for local material promotion. J. Afr. Earth Sci. 55, 69–
Geochim. Cosmochim. Acta 70, 1097–1112. 78.
Marques, R., Prudêncio, I.M., Rocha, F., 2011. Patterns of rare earth and other trace Nguetnkam, J.P., Kamga, R., Villiéras, F., Ekodeck, G.E., Razafitianamaharavo, A.,
elements in different size fractions of clays of Campanian-Maastrichtian Yvon, J., 2011. Alteration of Cameroonian clays under acid treatment.
deposits from the Portuguese western margin (Aveiro and Taveiro Comparison with industrial adsorbents. Appl. Clay Sci. 52, 122–132.
formations). Chem. Erde-Geochem. 71, 337–347. Njoya, A., Nkoumbou, C., Grosbois, C., Njopwouo, D., Njoya, D., Courtin-Nomade, A.,
McDonough, W.F., Sun, S.-S., 1995. The composition of the Earth. Chem. Geol. 120, Yvon, J., Martin, F., 2006. Genesis of Mayouom kaolin deposit (Western
223–253. Cameroon). Appl. Clay Sci. 32, 125–140.
McLennan, S.M., 1989. Rare earth elements in sedimentary rocks: influence of Rahman, M.J.J., Suzuki, S., 2007. Geochemistry of sandstones from the Miocene
provenance and sedimentary processes. In: Lipin, B.R., McKay, G.A. (Eds.), Surma Group, Bengal Basin, Bangladesh: implications for provenance, tectonic
Geochemistry and Mineralogy of Rare Earth Elements. Mineral Society of setting and weathering. Geochem. J. 41, 415–428.
America, vol. 21, pp. 169–200. Sawyer, E.W., 1986. The influence of source rock type, chemical weathering and
McLennan, S.M., Hemming, S., McDaniel, D.K., Nanson, G.N., 1993. Geochemical sorting on the geochemistry of clastic sediments from the Quetico
approaches to sedimentation, provenance, and tectonics. In: Johnsson, M.J., metasedimentary belt, Superior Province, Canada. Chem. Geol. 55, 77–95.
Basu, A. (Eds.), Processes Controlling the Composition of Clastic Sediments: Shimizu, H., Masuda, A., 1977. Cerium in cherts as an indication of marine
Boulder, Colorado. Geological Society of America, Special Publications 284, pp. environment of its formation. Nature 266, 346–348.
21–40. Singh, P., 2009. Major, trace and REE geochemistry of the Ganga River sediments:
Molua, E.L., 2006. Climatic trends in Cameroon: implications for agricultural influence of provenance and sedimentary processes. Chem. Geol. 266, 242–255.
management. Clim. Res. 30, 255–262. Singh, P., Rajamani, V., 2001. REE geochemistry of recent clastic sediments from the
Ndjigui, P.-D., Badinane, M.F.B., Nyeck, B., Nandjip, H.P.K., Bilong, P., 2013. source area weathering and sedimentary processes. Geochim. Cosmochim. Acta
Mineralogical and geochemical features of the coarse saprolite developed on 65, 3093–3108.
orthogneiss in the SW of Yaoundé, South Cameroon. J. Afr. Earth Sci. 79, 125– Skolkovitz, E.R., 1988. Rare earth elements in the sediments of the North Atlantic
142. Ocean, Amazon delta, and East China Sea: reinterpretation of terrigenous input
Ndjigui, P.-D., Bilong, P., Bitom, D., 2009. Negative cerium anomalies in the saprolite patterns to the ocean. Am. J. Sci. 288, 236–281.
zone of the serpentinite lateritic profiles (Lomié ultramafic complex, SE Taylor, S.R., McLennan, S.M., 1985. The Continental Crust: its Composition And
Cameroon). J. Afr. Earth Sc. 53, 59–69. Evolution. Oxford Blackwell, London, p. 312.
Ndjigui, P.-D., Bilong, P., Bitom, D., Dia, A., 2008. Mobilization and distribution of Temdjim, R., Boivin, P., Chazot, G., Robin, C., Rouleau, E., 2004. L’hétérogénéité du
major and trace elements in two weathering profiles developed on manteau supérieur à l’aplomb du volcan de Nyos (Cameroun) révélée par les
serpentinites in the Lomié ultramafic complex, South-East Cameroon. J. Afr. enclaves ultrabasiques. C. R. Géosci. 336, 1239–1244.
Earth Sci. 50, 305–328. Tchameni, R., Pouclet, A., Penaye, J., Ganwa, A.A., Toteu, S.F., 2006. Petrography and
Nesbitt, H.W., 1979. Mobility and fractionation of rare earth elements during geochemistry of the Ngaoundéré Pan-African granitoids in Central North
weathering of a granodiorite. Nature 279, 206–210. Cameroon: implications for their sources and geological setting. J. Afr. Earth
Nesbitt, H.W., Young, G.M., 1984. Prediction of some weathering trends of plutonic Sc. 44, 511–529.
and volcanic rocks based on thermodynamic and kinetic considerations. Toteu, S.F., Van Schmus, W.R., Penaye, J., Michard, A., 2001. New U–Pb, and Sm–Nd
Geochim. Cosmochim. Acta 48, 1523–1534. data from North-Central Cameroon and its bearing on the pre-Pan-African
Ngon Ngon, G.F., Etame, J., Ntamak-Nida, M.J., Mbog, M.B., Maliengoue Mpondo, history of Central Africa. Precambr. Res. 108, 45–73.
A.M., Gerard, M., Yongue-Fouateu, R., Bilong, P., 2012. Geological study of Zhou, H., Peng, W., Pan, J., 2004. Distribution source and enrichment of some
sedimentary clayey materials of the Bomkoul area in the Douala region (Douala chemical elements in sediments of the Pearl River Estuary, China. Cont. Shelf
sub-basin, Cameroon) for ceramic industry. C. R. Geosci. 344, 366–376. Res. 24, 1857–1875.