Dynamics of the transport of total suspended sediments in the Mefou watershed at Nkolbisson-Yaounde

The galloping demography in most of the urban centres of Cameroon poses real problems of urbanization both in terms of space management and in terms of essential resources supply. The management of the drinking water supply has become a very complex issue for many decades in the city of Yaounde, the capital of Cameroon. The supply of drinking water to communities in urban areas is usually done by building dams on different rivers. These dams create reservoirs to ensure the availability of the resource for drinking water supply, irrigation, and hydroelectricity production. The Mefou Dam, located at Nkolbisson in the western area of the city of Yaounde, filled in with water in 1969, is an example of a structure intended for the supply of drinking water. Generally, the construction of hydraulic structures often raises certain issues like the environmental impacts (Mc Allister et al. 2001), the economic facet (Barbier et al. 2009), and above all the sustainability aspect (WCD 2000).

The sustainability issue, which deals with the dynamics of flows and suspended solids, has been the subject of several studies in different countries of sub-Saharan Africa. These include Kattan et al. (1987) with the studies of the suspended sediment load and mechanical erosion in the Senegal basin; Boeglin & Probst (1996) with the studies of the transport of dissolved solids in a watershed of Niger. The studies of material transport in Cameroon began in the 1950s with the start of the research focused on the country's development and nature protection. It was initially undertaken on rivers to characterize the solid transport regime and to establish the mass balance of continental erosion at the country scale (Ndam Ngoupayou et al. 2014). Other studies about sediment dynamics in Cameroon are found in Olivry (1977), Nouvelot (1969, 1972), and Olivry et al. (1974). The case of the Mefou Dam, built on the Mefou River, located in a peri-urban area of Yaounde and providing 50,000 m³/day to the city potable water network is reported in this study. The activities in the area include deforestation, farming, and construction work, making the soil to be more and looser. So, soil materials that have become mobile are set in motion by the running water towards the weak depressions including the rivers. This large quality of sediments drained into the lake cause its siltation, thus avoiding it to fulfil its role of water storage for drinking purposes.

The objective of this work is, therefore, to quantify and qualify the solid materials which circulate into the river Mefou by suspension throughout a complete hydrological cycle. This information would clarify the dynamic of the degradation of the watershed and an indication of the rate of dam siltation.

The equipment used in this work was varied and depended on the tasks to be performed. The following main materials were made available by the Hydrological Research Centre of Nkolbisson in Yaounde:

• A Moulinet set: used to measure the discharge rate of water in rivers.

• A river gauge, was installed on the main watercourse in the locality of Etoud in order to record the daily water level.

• A cooler for conditioning the various samples taken.

The methodology applied here combines hydro-climatological data, quantification of Total Suspended Sediment (TSS), and processing of satellite images to elucidate the dynamics of solid transport in the Mefou River upper catchment. These include (1) monthly sampling of surface water and measurement of liquid discharges; (2) filtration of the various samples in the laboratory in order to quantify their concentration of suspended solids; (3) collection of rainfall data from the Hydrological Research Centre, (4) quantification of the solid component of the various flow fluxes using empirical formulas adapted to rivers as described by Sigha-Nkamdjou et al. (1997); and finally (5) deduction of sediment transport dynamics following land use patterns and water management at the Mefou Dam.

The watershed of the Mefou is subject to an equatorial climate of the Guinean type with four seasons of unequal importance (Olivry 1986; Suchel 1987). The interannual average rainfall is 1,668.60 mm (rainfall data from 2019 to 2021 of the Hydrological Research Centre of Nkolbisson, temperature data from the meteorological station of Yaounde). The monthly interannual average temperature is between 22.4 and 25.7 °C. They vary very little around 24 °C with annual thermal amplitude estimated at 2.4 °C (Kouam Kenmogne et al. 2013). The monthly average evapotranspiration is 82 mm in June and 113.2 mm during December, January, and February.

Field campaigns were used to carry out river flow rate, sample collection, and daily hydrometric monitoring.

The gauging sections were chosen at three locations (stations P1 of Ebong-Assi, P2 of Ozom, and P3 their confluence located at Etoud). Staff gauges were installed at P3 for hydrometric monitoring (Figure 3). Four river discharge measurements using the Moulinet devices were carried out on each of these sites, respectively, on August 10, 2020, October 10, 2020, January 09, 2021, and May 13, 2021.

The samples were taken at the surface at each gauging station (see Figure 3), using polyethylene bottles of 1 L each, which had been previously washed with ultra-pure acid and rinsed with distilled water. A total of 12 samples were taken per site (six during the dry season and six during the rainy season). Samples were adequately labelled and codified. The collected samples were transported to the laboratory for analysis.

Laboratory work consists of data review, sample filtration, and processing and interpretation of river discharge data. These analyses were carried out at the Soil Analysis and Environmental Chemistry Research Unit (URASCE) Laboratory of the Faculty of Agronomy and Agricultural Sciences (FASA) of the University of Dschang (Cameroon).

A particle of solid matter remains in suspension if the lifting force exerted on the grain is equal or greater than its weight, which makes it possible to know the speed limit from which a particle is suspended (Nouvelot 1969). There are several methods for estimating solid discharge. The method by double integration of the concentrations over the depth and the width of the watercourse is used here.

Calculations of material flows in Cameroon have been the subject of several studies using different methods (Nouvelot 1972; Olivry 1977), but also elsewhere (Walling & Webb 1981; Kattan et al. 1987). The choice of a method depends on the number and frequency of observations as well as their nature (Sigha-Nkamdjou 1994). In general, the calculation of the flow of materials is more precise as the sampling frequency is high and regular.

Fa (tons) is the total fluxes; Fa′ (tons) is the average monthly flux; Q_s (g/s): is the solid discharges; n is the number of measurements; N is the number of days in the month considered.

Sp (T/km²) is the specific degradation of the watershed; Fa (tons) is the total fluxes; A (km²) is the area of the watershed.

In Table 1, the water surface increases from 0.11% in 1987 to 0.13% in 2018, due to the widening of the beds of the watercourses first, and the redevelopment of the spaces occupied by the latter over the years (this is mainly Lake Mefou and the Lac Municipal which are the two main water reserves in the city of Yaounde in 2018 when this satellite image was taken).

During the 1-year observation period, four flow measurements were carried out on each of the representative rivers (Table 2).

The calibrating curves resulted from these data were used to estimate the liquid discharge the other days according to the water heights of the rivers. The average monthly values for the main river are presented in Table 3. As a result, variations of the flows on the different rivers are similar. Indeed, in these different watersheds, the flows are low during the dry seasons and relatively high during the rainy seasons, hence the presence of two maxima and two minima on the liquid flow evolution diagram.

The results of the concentrations of suspended solids of samples and different solid flow rates are given in Table 4.

Analysis of different samples of water was useful to determine the variations in TSS contents during the investigated period and estimate the average monthly solid discharges. Lowest values in the tributaries (4.47 and 2.25 mg/s, respectively, in Ebong-Assi and in Ozom) are observed in the month of August, which corresponds to the end of the short dry season. However, the lowest value (14.85 mg/s) is observed in the main river during the month of June after the estimate. Indeed, June belongs to the period of minimum water availability during the short dry season, a season during which there are a few rare rains which allow the materials to remain fixed to the soil by the roots of the plants which are developed. This is the cause of low suspended solids concentrations in rivers. It is noted that the solid flows almost evolve proportionally to the liquid flows. The solid discharge curves show the highest values in the rainy season and the lowest values in the dry season in the Mefou watershed.

The integration of monthly solid discharges over time has helped to determine the fluxes of solid materials that are drained by the rivers as a function of time. The results are shown in Table 5.

The results show that the solid transports in different rivers are always higher during the periods of higher flows than during the periods of lowest flows. September and October as well as April and May are generally the months with the highest flows, regardless of the watershed. By comparison, the river draining the East collector (Ebong-Assi) conveys a greater value of solid materials than the West collector (Ozom). The main Mefou River carries the highest quantities of materials with a total flux of 5.74 T/year. The fluxes found in the main Mefou watershed correspond to a specific degradation of approximately 75.89 × 10⁻³ T/km².year.

Four discharge measurements were carried out according to the seasons. The results obtained are roughly explanatory of the periods when these measurements were made. Indeed, these results combined with the daily levels recorded on the Mefou River, upstream of the lake, attest that liquid flows are higher in the rainy season than during the dry season. The differences observed in the results of the East and West tributaries and on the main river are due to various parameters such as the shape and area of the basin, the land use, and the soil type. In the East and West sub-watersheds, respectively, at Ebong-Assi and Ozom, as in the Mefou watershed, the months with the highest average monthly liquid flows are September and October during the long rainy season and April and May during the short rainy season. This justifies the Ombrothermic diagram of Yaounde (2019–2021) where maximum precipitation is observed during these periods. This confirms the findings of Sighomnou (2004) who declared that the monthly variations in flow follow the shapes of the characteristic histograms of the regions presented.

During the rainy months with the highest flows, the liquid flows have values of 1.507 m³/s in October, and 1.139 m³/s in May in the eastern catchment basin, against 1.431 m³/s in October and 0.714 m³/s in May in the western collector watershed. Similarly, during the driest months, these liquid flows are 0.333 and 0.179 m³/s, respectively, in the East and West in January, while they are 0.447 and 0.255 m³/s in August. By way of comparison between the two collectors, the East has higher liquid flow than the West. These differences can be explained by the fact that these two rivers represent different environments and undergo different natural and anthropogenic actions. But also, the river located to the East meets more tributaries and has a wider flow channel. By way of comparison between the two collecting rivers, knowing that the origins of the waters are identical, that located to the East has a higher liquid discharge than that located to the West, because, with respective areas of 36.19 and 24 .08 km², the East River has a greater number of tributaries and therefore a higher probability of mobilization of fluid elements. Indeed, Figure 3 shows that the ECW has a main watercourse of order 3, while in the WCW, the main watercourse is of order 2.

The estimated liquid discharges in the Mefou River at Etoud vary from 1.93 to 3.92 m³/s in the rainy season against 0.58–0.80 m³/s in the dry season. Kouam Kenmogne et al. (2013) gives different results on this same river at its outlet at the entrance to the lake. In fact, they vary between 1 and 1.2 m³/s in the dry season and between 1.7 and 2.2 m³/s in the wet season. The discrepancy between these results can be attributed to the time span of investigations. While the present holds on estimated results based on a complete hydrological year, the former work investigation period was much shorter. It should also be noted that in the rainy season, the evacuation valves of Lake Mefou are constantly open on the dam to avoid overflows during floods. These openings thus maintain high water flow velocities in the tributaries during the rainy season. In addition, Kouam Kenmogné (2013) in this same watershed gave a maximum liquid discharge on this river at 4.14 m³/s in October. This result is approximately the same as the 4.124 m³/s measured also in October during the present study. This similarity could be explained by the measurement methods, but also by the fact that they have been taken during the same hydrological period. Indeed, these two studies use the same method and cover almost the same period.

In the East collector at Ebong-Assi, the monthly solid discharges are about 226.05 mg/s in October during the long rainy season and 28,62 mg/s in the West collector at Ozom. It is the same throughout the study period where the solid of eastern collector data are always more important than those of the western collector. These two sub-watersheds rest on the same base, thus giving them the same physical parameters of the soil. The fact that the eastern watershed has a larger surface area, the rivers are denser and with more ramifications and the demography is greater (and therefore greater anthropic activity) than the western one. It could justify that the quantities of solid materials transported in the East River are greater than those from the West River.

Compared to the other results, Mekok (1999) obtained on the Sonkwé river in the village of Nkoékouk South of Cameroon, an annual flux of 13.77 T/year, while they are, respectively, 84 T/year (in 1996) and 163 T/year (1997) on the Nyangoong River, 336 T/year (1996) and 211 T/year (1997) on the Minwo River in Enguepanyou village. For Tamonkeng (2003) on the Kienké River in Kribi in the South region, these flows are 28,126 T/8 months. These four watersheds belong, like the two of Ebong-Assi and Ozom, to the same climatic context, that of the Guinean equatorial type. The differences observed in these results could be justified firstly by the difference in their area, but also by the fact that they do not have the same geographical and geological characteristics. Geologically, the Mefou watershed is located on the northern edge of the Congo craton, while the other watersheds are located in the South Cameroon domain of the Pan-African North Equatorial Range in Cameroon. This geological difference gives them different physico-chemical and mechanical properties of the soil, and therefore they behave differently concerning the erosion process. In terms of geography, South Cameroon is a region that is less subject to anthropogenic influence than the central regions because of industrialization and the migration of populations to the latter, in search of well-being.

In the Mefou River at Nkolbisson-Yaounde, the TSS fluxes during the study campaign from June 2020 to May 2021 are estimated at 5.74 T of materials in an area of 75.62 km². These results are different from those obtained by Ndam Ngoupayou (1997) during the campaign of March 2003 to February 2004 year in the Nyong watershed in Mbalmayo (located South of Yaounde). The TSS fluxes were estimated at 49,100 T of materials covering an area of 13,555 km². The difference between these two cases could be explained by the fact that TSS in the Mefou watershed are drained by an average liquid discharge of 1.64 m³/s, very small compared to the 127 m³/s in the Nyong River. Also, the watershed of the Nyong covers a very high population, which causes great anthropic activity, and its large surface is at the origin of a greater mobilization of solid elements towards the rivers of low funds compared to that of the Mefou watershed.

The specific degradation is 75.89 × 10⁻³ T/km²/year in the Mefou watershed (75.62 km²) for a total quantity of 5.74 T/year of materials transported. Olivry & Hoorelbecke (1975) study carried out in the Far North region of Cameroon obtained 7,130 T/year of the sediment flow in the upper basin of the Mayo Tsanaga (25.46 km²), for a specific degradation of 280 T/km²/year. While these two rivers have each a dam of the same nature the low values observed in the Mefou watershed would be due to the fact that the thrusting and saltation not having been studied, only the suspended solids constitute the solid materials, contrary to the Mayo Tsanaga watershed.

The dynamics of TSS transport were carried out in the upper Mefou watershed with the aim of highlighting soil erosion according to the seasons. Fieldwork and laboratory analyses enabled to collect adequate data. As a result, liquid discharges are linked to the watershed behaviour according to the seasons, and are proportional to the heights of water of the rivers, which themselves depend on variations of the precipitations and therefore more generally on the climate. Solid discharges also show an evolution according to the seasons (abundance or not of mobilizable materials). The materials transported by suspension contribute to the filling of the lake and the reduction of its water storage capacity. They undergo the phenomenon of decantation at the level of the lake because of the loss of speed and the stagnation of water due to the dam. The trend is the same when looking at the results of the specific degradation of the watershed, where the materials are mostly mobilized during rainy periods. The fact that the main Mefou River carries a total flux of 5.74 T/year which corresponds to a specific degradation of approximately 75.89 × 10⁻³ T/km².year show that the dynamics of sediment transport in the Mefou watershed is not yet very accentuated. This could be due to the fact that it is an environment located in a peri-urban area, where the soils are still fairly protected by the plant cover and human activity is not very important.

However, these suspended solids could over time become an obstacle to the storage capacity of Lake Mefou. Thus, the protection of this space is necessary because of its capital importance for the populations, but also for better management of the polluting elements which flow into the lake.

In order to densify this study, it would be important to measure sediment transport in the whole catchment area and over a longer period. It should also be important to include materials drained by bedload and saltation during the analysis of the sediment transport.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.