Abstract: This study reports on the distribution of trace elements in the sediments of the Noun River valley. Two groups of trace elements have emerged based on their distribution and geochemical behaviour. The first group shows lower values whereas the second group aligns higher values. Elements like Al and Sr in one side and Zn, Zr in the other present concentrations statistically different from those of the rocks compared to other elements in trace analyzed (Cr, Ni, Ba, Co, Cu, Fe, Mn, V, As and Pb). The trace elements Al, Sr, Zr, Zn and Pb in the sediments of the Noun River show concentrations rather lower than those of the world. The average concentrations in elements C (3.72%), N (0.27%), S (0.05%) and of the ratios C/N (13.77) and N/S (5.29) reflect those resulting from the decomposition of the organic matter which derives from soils with very small concentrations in S. The Noun valley sediments are mostly of primary (quartz, microcline, plagioclase, orthoclase, ilmenite, anatase) and newly formed or secondary (kaolinite, gibbsite, hematite, goethite) minerals. This mineralogy and the overall low concentration of the studied trace elements, with Al, Fe, Ti and Mn as most abundant elements, is consistent with the local geology made up of the basic rocks (basalts, volcanic ash) and of the acid rocks (granites, gneiss). Overall the trace elements concentration and distribution suggest their geogene origin. Yet, this study is a first contribution of its kind towards the development of a baseline geochemical database essential for a sustainable and healthier Noun valley ecosystem.
`INTRODUCTION
Over the last decades, regional studies on the distribution of trace elements have been gaining much interest due to the established link between their deficit or excess in the environment and human related health. In this context, a geochemical study of river basin sediments can help understand and identify the sources of trace elements. Their total load in sediments depends on a myriad of factors including the local geology, human activities along river beds and the ecological health of the system. As it can be inferred from Gailardet et al. (2003), trace elements distribution in the sediments can be regarded as the signature of all physical, chemical and biological processes affecting an aquatic ecosystem. It is within this framework that regional management of sediments to provide the much needed baseline geochemical information has been getting interest world wide (Muller et al., 1996). This study was motivated by a probable environmental threat to the Noun valley from its active and growing population, economy and quasi intensive agriculture.
Furthermore, despite its relatively cardinal importance, the Noun River basin has not been the focus of a study on the distribution of trace elements in its sediments. Thus, the major objective of this study was to assess the distribution and fate controlling mechanisms of selected trace elements in the sediments of the Noun River and in some of its tributaries. It lays ground towards the development of a comprehensive geochemical database. Such a database is needed as guidelines and for monitoring trace elements cycle from the sediments to the food web.
MATERIALS AND METHODS
Study area: The studied area is part of the high plateaus of the Western Cameroon (HPWC) (10° and 11° of longitude East, 5° and 7° of latitude North) in the central part of the Volcanic Line of Cameroon (VLC). The average altitude of the HPWC is 1400 m. The Noun River drains much of the studied area. Its principal tributaries include the eastern rivers Mafoumba, Maouat, Temudjin, Panké and Nkoup whereas in the West flow the Mifi (Southern and Northern), Nkong, Ndé, Ngam, Mbougam and Bagam (Segalen, 1967). The climate is of the monsoon subequatorial type (pseudo tropical) of wet and fresh predominance with one rainy season from mid-March to mid-November. The annual precipitations are higher than 1400 mm. The temperatures are refreshed by altitude. Bafoussam records the maximum between 23 and 27°C with annual average of 20°C.
The geological formations of the HPWC consist primarily of neoproterozoïc base (Dumort, 1968). This base consists of varied granitoid (granites, granodiorites, syenites and diorites), gneiss and orthogneiss but also of volcanic rocks of the SW-NE trend line known as the "fracture" of Cameroon. These volcanic rocks are acidic (rhyolitic formations of Mbam, Mbapit, Nkogam, plain of Tikar, etc…) and also basic (andesites and basalts of varied forms including pyroclastics made up of ashes and particulates projections). In general, the overburden formations are primarily of volcanic nature. They consist of basalts to more than 80%, trachytes and of rhyolites and pyroclastics belonging to the line of Cameroon. The alluvia extend in a considerable way in certain valleys. Soils are young, hydromorphic and especially ferrallitic, fairly desaturated and of a good fertility which is related to the region volcanic activity (Segalen, 1967).
Sampling: The sediments sampling was carried during the rainy season in September-October 2003 on five points along the Noun River (160 km long of North-South flow direction) and on five other points in five of its tributaries (Mifi-South, Mifi-North, Nkong, Ngam, Nkoup) (Fig. 1).
The physicochemical parameters measured in situ include pH, electric conductivity and the temperature of water at each point of sampling. Sediment samples were collected in fine grains zones assumed trap of trace elements (Deschamps et al., 2002). They were collected with a stainless steel sampler at approximately 1 m of the river’s banks in sectors of reduced flow. Composite samples were made from samples collected 1 m apart along the river. The fraction lower than 2 mm obtained after passage through a sieve of 2 mm mesh was collected in polyethylene bags. It was ambient air dried and on constant weight by lyophilisation (Matschullat et al., 2000). Then the dry sieving of the samples was carried out through a nylon sieve of 63 μm (Borba et al., 2000; Matschullat, 2000). This sieving made it possible to obtain the fine elements (argillaceous fraction) most active physico-chemically and with high adsorption capacity. The samples were homogenized to minimize analytical uncertainties and were preserved in glass vials for analyses.
| Fig. 1: | Sediment sampling points |
Analytical procedures: The determination of trace elements was preceded by a total digestion of 0.1 g of the fraction lower than 63 μm using in 5 mL of 65% supra pure HNO3 and 3 mL 40% supra pure HF through Waldner apparatus. Solutions of samples obtained after filtration using FILTRAK® papers were analyzed by routine ICP/AES for the determination of selected trace elements including Al, Fe, Ti, Sr, V, Cu, Co, Cr, Ba, Zr, Ni, Pb, Zn, Mn). GF/AAS with a PERKIN ELMER 4100 ZL apparatus coupled with a FIAS 200 for hydride generation was used for As determination. Quality assurance and control procedures included comparison with reference samples. There is a good harmony (≥ 83% of recovery) between measured values and certified values. The determination of the concentration of carbon (C), nitrogen (N) and sulphur (S) was carried through elementary CNS analysis protocol following the dry method (Pansu et al., 1998) with the Elementar VARIO EL apparatus. The sediments mineralogy was determined by x-rays diffractometry and electronic microscopy (Blasig, 2004) on samples of granulometry lower than 63 μm.
Statistical procedures: Cluster analysis was used to identify affinities between the samples or elements whereas the Principal Component Analysis (PCA) assessed the rates of strict variance between samples or between trace elements. The correlation of the products of Pearson moments are also studied for the affinities and the tests W of Mann Whitney (Wilcoxon) were used to compare medians between sediments.
RESULTS AND DISCUSSION
Relationships between the physico-chemical characteristics of water of the Noun River and its tributaries and the behaviour of trace elements: The quasi-neutrals pH from 6 to 7.5 (Table 1) of water of the Noun River and of its tributaries are in the range (6.5-8.5) of unpolluted natural surface water (Radojeviæ and Bashkin, 1999; Weiner, 2000). This observation suggests water with low mobility of metallic trace elements as opposed to acid mine water generally rich in trace elements and very dependent on protonation/ deprotonation reactions of the type:
Me2+ + 2 H2O ↔ Me (OH)
2 +2 H+ (Me denotes a metal) |
The electric conductivity observed in water of the Noun River and its tributaries (22.5-95 μS cm-1) is far below Bifano and Mogollón (1995) clean water of rivers (430 μS cm-1) and contaminated (950 μS cm-1). The observed values are rather contained in the range from 10 to 100 μS cm-1 given by Radojeviæ and Bashkin (1999) for surface water. These low values of electric conductivity are consistent with the quasi neutral pH values and the low contents in trace elements. These conditions of pH and electric conductivity contribute more towards the precipitation of elements like Fe2+, Fe3+, Mn2+, Al3+ with possibly the formation of complexes with ligands NO3¯, HPO42¯, H2PO4¯ which can precipitate. The electric conductivity which appears high with 95 μS cm-1 in the tributary sample S8I7 (Nkoup) may be explained by fact that this tributary drains the volcanic ash rich in Ca2+, Na+, Mg2+, SO4¯. Besides, the temperatures recorded in water (20.4-24.7°C) reflect only the natural and normal process which characterizes the climate of the area.
| Table 1: | Physico-chemical parameters of water |
**Samples from Noun River |
|
| Fig. 2: | Comparison of Al, Ni, Zn and Cr concentrations in sediments
along a flow path |
Influence of water flow on the distribution of the trace elements along the Noun River and its tributaries: Some trace elements such as Al, Ni, Zn and Cr were selected to assess the dilution phenomenon in the sediments. Their concentrations along the sampling flow path are compared (Fig. 2) and also a basis for discussing their behaviours taking into consideration their physicochemical properties (for example geogene and anthropogene and the difference in their mobility).
Aluminium: The Al concentrations along the Noun and in the tributaries are quasi-constants. This constancy is probably related to Al tendency to immobility (Reimann and De Caritat, 1998), which is consistent with the near neutral pH of the Noun water of 6-7.5 less favourable to the solubility of the majority of natural Al compounds (Weiner, 2000). Thus, the low concentrations of Al reflect those originating naturally by the degradation of the acid rocks (granites and gneiss) or basic (basalt) rocks of the area. These types of rocks present the lowest Al variations as they contain respectively 73,000 mg kg-1 in the granites and 83,000 mg kg-1 in basalts (Reimann and De Caritat, 1998).
Nickel: The variations of Ni along the Noun agree with its mobile character. Weiner (2000) argues that Ni is one of the most mobile metals of the aquatic environment. The increase in its concentration in the Noun sample S7J7 (road Noun-project of Noun) may be most likely related to the pH. It can also be linked to the decrease of the hydraulic gradient and the resulting prevalence of precipitation/co-precipitation process at this sampling location (altitude of 945 m). This explanation also applies for the high Ni concentration observed in the sample S1D5 (Bamendjing). The neutral water pH 7 combined with this sampling point located just downstream from the dam where the local hydraulic conditions might be prone to metal accumulation. Ni lowest concentrations which appear in the samples from the tributaries S5J6 (Nkong), S8I7 (Nkoup), S6J7 (Ngam) are undoubtedly related to the drainage by these tributaries of the granitic and gneissic areas where the rocks have lower Ni concentrations.
Zinc: Zn and Ni have similar behaviour. They are observed in higher concentrations in samples S1D5 and S7J7. The slight variation of the concentrations in the other samples along the Noun can be explained by Zn mobility as Zn2+ species under oxidizing conditions of running and slightly acidic water (Reimann and De Caritat, 1998). These concentrations along the Noun River seem mainly influenced by the location of the sample S2 E3 (Mifi-North) because of its almost similar altitude to that of S1D5 (Bamendjing). The low concentrations observed in the tributaries are also probably related to the drainage of the local geology materials of low Zn concentrations.
Chrome: The low mobility character of Cr would explain the slight variation of its concentrations observed in the samples taken along the Noun River, except the sample S1D5 which exceptionally presents a high concentration. Similarly lower concentrations of Cr and Ni found in the samples of the tributaries such as S5J6 (Nkong), S8I7 (Nkoup), S6J7 (Ngam) can be due to their natural association and common origin. In general, the behaviours of Al and Ni, Zn, Cr are almost divergent. Aluminium seems not to be influenced by the contributions of water of the tributaries because its concentrations remain almost identical in the various samples contrary to three other elements Ni, Cr, Zn which present a slight variation along the Noun. Dilution does not seem to be an important factor for the three elements, because the sample S7J7 (Noun-road project of Noun) from a lower relief location presents even higher concentrations in these elements. The statistical analyses carried out for these samples as shown it the dendrogramme (Fig. 3a) and the principal component analyses (Fig. 3b) confirm these similarities in the composition of the samples in trace elements along the Noun and the divergences between these samples. Also, the samples along the Noun present a very great affinity between them suggesting that they are not affected by the dilution phenomenon. The only sample which presents a very great variation in its composition is S1D5 ** (Noun-Bamendjing). The difference which exists between this sample and the others is probably related to the fact that it comes from a sampling point located just downstream of the Bamendjing dam. This sampling point acts as a sink for heavy metals due to the near neutral pH and quasi stagnant water that favours precipitation and co-precipitations of dissolved heavy metals.
Appreciation of the natural background of trace elements in the sediments: The term background means here the absence of geochemical anomaly. The evaluation of the sediments background can be achieved through comparison with local geology and to the sediments of the world.
Total contents of trace elements in sediments and rocks: The total concentrations of trace elements in the sediments of the study area are compared to those of the rocks of the HPWC (Table 2) as reported by Fosso (1999), Kamgang (2003), Talla (1995), Tchouankoue (1992) and Wandji (1995). Statistically significant differences of certain elements (Al, Zn, Zr, Sr) with 95 and 99% of confidence limits indicate values of Al and Sr lower than those of the rocks whereas Zn and Zr higher than those of the rocks.
| Table 2: | Geochemical composition of sediments of the Noun River and
its tributaries compared to estimated mean values in the rocks of HPWC.
All reported values are in mg kg-1 dry weight of sediments and
rocks |
*Elements showing a statistically signicative difference.
Mean values of rocks (Fosso, 1999; Kamgang, 2003; Talla, 1995; Tchouankoue,
1992; Wandji, 1995) |
|
| Fig. 3a: | Grouping of samples as function of their composition |
| Fig. 3b: | Discrepancies in sample composition |
The remaining elements (Cr, Ni, Ba, Co, Cu, Fe, Mn, V, As, Pb) have similar concentrations with those of the rocks. The observed differences can be related to the elements intrinsic properties, the mechanisms governing their accumulation or even the mechanical sedimentary differentiation, which can increase their concentration.
Aluminium and strontium: These elements have rather higher concentrations in rocks than in the sediments. The median values presented for Al are respectively 141.600 and 116.367 mg kg-1 for the rocks and the sediments. These differences for the two types of samples can be due to the immobility prone character of aluminium coupled with the near neutral pH (6-7.5) of the draining water and hence its low solubility. The amount of aluminium which is found in the sediments can be related to normal sedimentation because it does not show any positive correlation (Table 3) with elements likely to support its accumulation.
Sr presents the respective median values of 569.500 mg kg-1 for the rocks and 147 mg kg-1 for the sediments. It does not present any correlation with pH, C and minerals in the sediments. Its low concentration in the sediments can be related to its similar geochemical characteristics to those of calcium and magnesium. These elements can through competition take over the adsorbing sites in the sediments and consequently Sr remains much mobilized by water.
Zinc and zirconium: Zn and Zr have median concentrations higher in the sediments than in the rocks. Zn respectively presents median concentrations values of 81 mg kg-1 in rocks and of 142 mg kg-1 in sediments. Zn accumulation may be through the mechanism of precipitation/coprecipitation from water draining the studied area. It presents positive correlations with the pH of water (0.55), Hematite-Zn (0.65) and Gibbsite-Zn (0.52) (Table 3).
| Table 3: | Pearson correlation of some trace elements in the sediments |
* Correlations not less or equal to 0 statistically significant
at confidence level of 95% |
|
The coprecipitation of Zn might have occurred through minerals like gibbsite. It is favoured by the near neutral pH of water as supported by the positive correlation between the gibbsite and the pH (0.58). Zn is also probably retained through adsorption on iron oxide surfaces such as hematite.
Zr shows concentrations of 222 mg kg-1 in rocks and of 398 mg kg-1 in sediments. The observed higher concentrations in the sediments is probably due to the mechanism of coprecipitation taking considering its very low solubility under the pH range of 6-7.5. As for Zn, coprecipitation with gibbsite under the influence of the pH may have taken place as supported by the positive correlations of Zr-pH (0.61) and gibbsite-Zr (0.86). Besides, the dendrogrammes (Fig. 4a) and the correlations of Pearson (Table 3) suggest the following natural associations: Al and As; Mn and Pb; the majority of transition metals Ni, Co, Cr, Cu, Fe, V; Zn, Ti, Zr and that of Ba, Sr.
These regroupings can thus have been favoured by the natural origin of these elements but also by their intrinsic properties because of their similar behaviours under certain conditions. Besides, the concentrations of the analyzed trace elements confirm their order of abundance in the earth's crust and in the parent materials (rocks of the area). This relative abundance is consistent with the principal component analysis showing the percentage of variance of trace elements in the sediments (Fig. 4b). These elements appear in high concentration according to the order: Al > Fe > Ti > Mn.
Trace elements in the noun sediments versus the sediments of the world: The comparisons between trace elements in the sediments of the studied area and those of the world are presented (Table 4). The elements Al, Sr, Zr, Zn which previously were found with variations compared to the concentrations of the rocks are reported in the range of the sediments of the world presented by Deschamps et al. (2002) and Reimann and De Caritat (1998).
| Fig. 4a: | Grouping of trace elements |
| Fig. 4b: | Divergence of trace elements in samples |
It is the same for Pb. However, certain elements like Ni, Ba, Co, Cu, Fe, Ti, V present concentrations higher than the range found in the world. This variation can be due to many factors in particular those related to the inherent differences on the rocks, minerals and soils which were drained in the sediments analyzed in the world. The list of factors may also include the size of the grains used for the analyses and the type of analyses carried out. For example, Ti concentration in the sediments of the studied area (18.920 mg kg-1), twice larger than the maximum of concentration met in the world, can be due to the fact that the studied area presents minerals rich in titanium.
| Table 4: | Comparison between the geochemical composition of the sediments
of the Noun River and its tributaries with estimated mean values of the
world. All reported values are in mg kg-1 dry weight of sediments |
a: data compiled in Deschamps et al. (2002); b: data
compiled in Reimann and De Caritat (1998) |
|
Carbon, Nitrogen and Sulphur (CNS) contents and the sediments mineralogy: The CNS content has been shown by Matschullat et al. (2000) as important for assessing a geochemical background. The sediments mineralogy is the signature not only of the recent geological history but can also indicate the potential anomaly of trace elements distribution.
CNS: The sediments ratios C/N (13.77) and N/S (5.29) and the slightly low values in particular for C (1.84-5.74%) with average of 3.72%; N (0.13-0.42%) with average 0.27% and for the S (0.03-0.06%) with average 0.05% suggest that the elements CNS in the sediments originate only from the weathering of the soils. These results are almost similar to those of Evans et al. (1995) in the study of the sediments of Ensenada de San Simón in Spain with values of C (0.4-6.5) with average 4.8% and N (0.1-1.8) with average of 0.4%. These values and their corresponding C/N ratio of 2.4-13.2 with average 10.7 were explained as signature of the predominance of the organic matter from soils. In general as noted by Meybeck (1993), the concentrations in CNS in the sediments reflect the conditions of degradation, thickness of the soil transported by water, climate and even also of the type of vegetation of the area (savanna). The N content of approximately the 1/14 of C and the S content the 1/6 of N suggest that these CNS elements could be a continuity of the biogeochemical cycle in the sediments, with however moderate concentrations. Lastly, the very low content of S can characterize the sediments which do not result from the acid sulphate soils.
| Fig. 5: | Abundance of minerals in the sediments |
Values of S less or equal to 0.10% and of C superior to 4% in the sediments are interpreted by Österholm and Åström (2002) as not resulting from the acid sulphate soils.
Mineralogy: The minerals observed in the sediments of the studied area (Fig. 5) are mostly primary quartz, microcline, plagioclase, orthoclase, ilmenite, anatase and of newly formed kaolinite, gibbsite, hematite and goethite.
This mineralogy corresponds with the regional geology characterized by the prevalence of the pluto-volcanic and metamorphic acids (granite and gneiss) and basic (basalts, ashes and lapilli) rocks, as well as the overlying soils. In general, the observed minerals do not indicate some geochemical anomalies that could be related to trace elements. They are rather common products of intense chemical weathering of tropical regions with ultimate degradation products such as kaolinite and gibbsite (Johnsson, 2000). Mention should however be made on the presence of titanium minerals ilmenite (FeTiO3) and anatase (TiO2), which probably contributed to the relatively high concentration of Ti in the sediments. Similarly, the abundant presence of silicates and iron oxides such as ilmenite may explain the relative high concentrations of Al and Fe in the sediments.
CONCLUSIONS
The concentrations of trace elements in the sediments of the Noun River and its tributaries reflect those of the acidic and basic rocks of the region with Al, Fe, Ti and Mn as the most abundant elements. The mechanisms governing their distribution include their intrinsic geochemical properties; precipitation and coprecipitation, as well as adsorption on Fe (hematite), Mn (mainly for Zr, Zn, Pb, Cr) oxides and Al hydroxides (gibbsite), kaolinite and the organic matter surfaces (likely for Cu). The influence of the dilution which would have resulted from the rainy season related leaching seems negligible. The sediments downstream in the vicinity of the Bamendjing dam is the only sampling location with divergent concentrations of trace elements. The variations observed with ranges found in the world are due to several reasons that may include slight differences in soil and minerals types, the size of the grains used for the analyses or also the types of analytical procedures carried out. The observed minerals reflect those generally found in the tropical areas. However, the presence of Ti bearing minerals mainly ilmenite and anatase are probably the main source of its relatively higher concentration. The composition in biogenes elements C, N and S can only be the continuation of the biogeochemical cycle in the sediments. Overall this study is a first approach and a contribution for the establishment of the environmental levels of trace elements in the sediments of the Noun River and its tributaries. It lays grounds towards the development of a geochemical baseline of data essential for the sustainable development of the important ecosystem of the Noun valley.
ACKNOWLEDGMENTS
The first author acknowledges the financial support from the German Academic Exchange Service (DAAD) for the fieldwork in Cameroon and the living expenses at the Technische Universität Bergakademie Freiberg in Germany.