Evaluation of vulnerable zones to water pollution in the lower Soummam alluvial aquifer, Bejaia, Algeria

In Algeria, water resources are exposed to very significant industrial and urban discharges (Mebarki, 2000), caused by the rapid acceleration of industrialisation and urbanisation. These anthropogenic inputs can contribute to the chemical composition of the aquatic environment, such as high concentrations of harmful contaminants like nitrates (Abbasnia et al., 2019; Zhai et al., 2019; Liu et al., 2021).

In addition, since the 1990s, Algeria has been suffering from a major drought due to erratic rainfall in time and space, which has worsened in the last three years. As a result, there are many concerns about the availability of water in terms of quantity and quality (Ghenim & Megnounif, 2013); worldwide water demand is expected to increase by 20–30% by 2050 (Wada et al., 2016). In addition, intensive groundwater extraction and the discharge of untreated wastewater into watercourses make the aquifer vulnerable to pollution.

However, without adequate knowledge of the spatial distribution of major and trace ion concentrations, sustainable management of available groundwater resources cannot be achieved. This knowledge helps to understand the distribution of water quality parameters and the risk of vulnerability to pollution. It then enables the development of an overall concept for the management, protection and use of resources with a view to sustainable development (Vaudan et al., 2005).

To this end, the aim of this work is to contribute to a better understanding of the process of increasing water mineralisation through sampling and analysis campaigns during high and low water periods. This was done in order to identify areas vulnerable to water pollution in the river and in the alluvial aquifer of the lower Soummam. This work will therefore shed light on the nature of the exchanges between the river and the alluvial aquifer, the filtration mechanisms and the properties of the contaminants that govern the exchanges between the alluvial aquifer and the river. It will also make it possible to monitor the physico-chemical constituents of the water in order to guarantee its quality. This water quality study is very important because of its role in preventing waterborne diseases and identifying sources of pollution in the region. This is despite the fact that Algerian legislation has an arsenal of laws and regulations, whose effective implementation on the ground remains inadequate.

In the literature, there is evidence of the existence of several works in which multivariate statistical techniques have been used in the assessment of water quality in order to understand hydrogeochemical variation and also the sources of pollution (Alves et al., 2018; Rakotondrabe et al., 2018; Andrade Costa et al., 2020). In fact, there are two types of hydrogeochemical approaches used to characterise surface and groundwater. The first is a study of the characterisation of groundwater and surface water based on a large amount of physico-chemical data (André et al., 2005; Golchin & Azhdary Moghaddam, 2016). The second allows the monitoring of the spatial and temporal evolution of physico-chemical parameters and the estimation of their origin through their correlation with the geology and piezometry of the alluvial aquifer (Gibbs, 1970; Vázquez-Suñé et al., 2005; Tay et al., 2017).

In the literature, work has been done on the use of hydrogeochemical tracers and the analysis of physico-chemical constituents to study exchanges between rivers and aquifers (Kessasra et al., 2021; Ghodbane et al., 2022). These studies show that the exchange between rivers and aquifers is a complex phenomenon that can be influenced by many factors, such as the geometry of the interface zone between the river and the aquifer, the nature of the aquifers and the hydroclimatic conditions. However, the methods used are effective and allow a better understanding of the complex mechanisms that govern these exchanges and a better identification of areas vulnerable to pollution.

Building upon existing knowledge, this study leverages findings from past research on alluvial aquifers (Bencer et al., 2016; Djemai et al., 2017; Mirzabeygi et al., 2017) to address critical gaps in the understanding of Alluvial Aquifer in the Lower Soummam, Bejaia, Algeria. We aim to achieve this by the following:

Identifying vulnerable areas: by implementing a novel multi-phase approach with high and low water sampling campaigns combined with multivariate statistical analysis, we aim to pinpoint specific areas susceptible to pollution within the river and alluvial aquifer with significantly higher accuracy compared to previous research.

Elucidating spatiotemporal dynamics: employing hydrogeochemical tracers and analysing physico-chemical constituents during both high and low water periods will allow us to delve deeper into the complex exchange mechanisms between the river and the aquifer. This refined understanding will significantly improve the ability to predict and manage potential threats to water quality, contributing to long-term aquifer sustainability.

Developing a robust monitoring system: the comprehensive data collection network utilising spatially distributed boreholes, wells and stations will establish a robust monitoring system for the river's physico-chemical constituents.

This system will provide ongoing data crucial for implementing informed and dynamic water resource management strategies, preventing water quality decline and ensuring the aquifer's long-term viability.

This study is part of a project to characterise the waters of the lower Soummam Valley in Bejaia, Algeria. Its main aim is to establish the hydrogeological and hydrogeochemical context of the water by modelling the chemical interaction between surface water and groundwater. It has also verified and confirmed the infiltration and drainage between the river and the groundwater.

To better monitor flow variations in the river, we have installed three flow measurement stations along the main watercourse (upstream, downstream and an intermediate station). Two series of flow measurements by differential gauging were carried out in 2019 and 2020 during the low water period from June to September.

Monitoring the spatial and temporal progression of a series of parameters will allow us to understand the mechanisms by which mineralisation is acquired and the process associated with the geochemistry of the groundwater and surface water. This will allow us to determine the exchange between the river and the alluvial aquifer and to interpret the hydraulic relationship between them.

The study area selected is the watershed of the lower Soummam valley. It is located in the wilaya of Bejaia in north-eastern Algeria. The alluvial plain of the lower Soummam extends over a length of 45 km with a width of between 700 and 2,000 m and covers an area of 709 km² (Saou et al., 2012). It is characterised by a humid climate with relatively high rainfall, between 600 and 900 mm, with summer temperatures varying between 24 and 28 °C. This relatively high rainfall feeds the alluvial aquifer of the lower Soummam and Oued Soummam rivers.

To establish hydrogeochemical measurements, two water analysis campaigns were conducted. These waters come from the Quaternary alluvial aquifer in the lower Soummam. The first campaign coincides with the period of high water (flooding) during which the river receives a significant amount of water from its tributaries and the upper Soummam Basin. The second campaign coincides with the period of low water (receding) during which the river receives only the contributions of the upper Soummam Basin; during this period, the tributaries of the lower Soummam were dry at the confluences with the main watercourse and do not contribute any flow.

For groundwater, 25 water samples were taken from eight irrigation wells and 17 boreholes. Concerning surface water, seven water samples were taken from seven measuring stations distributed along the river from upstream to downstream. These stations were chosen to study the behaviour of the alluvial aquifer and the river using hydrogeochemical tracers during both periods. Water samples were taken under the surface in turbid areas where the water column is well mixed (VanTrump & Miesch, 1977; Lorite-Herrera et al., 2008). To eliminate stagnant water in the wells and boreholes, samples were taken as long as possible after pumping until electrical conductivity (EC) and pH had stabilised.

The University of Bejaia, in particular, Research Laboratory in Applied Hydraulics and Environment, has all the necessary equipment to analyse the water samples and subsequently obtain measurements of physical parameters, alkalinity, major elements, trace elements and flow rates. In fact, physical parameters (temperature, pH (potential of hydrogen), EC and dissolved oxygen (DO)) were measured using the portable multi-parameter EXSTIK II pH/conductivity EC 500 instrument; it is a portable, multi-parameter instrument packed with capabilities for measuring key water quality parameters directly at the point of interest. Alkalinity was measured by automatic acid–base titration CRISON. Many CRISON titrators offer high precision and automation, making them suitable for accurate alkalinity measurements in scientific research and various other applications.

Major elements (Ca (calcium), Mg (magnesium), Na (sodium) HCO₃ (bicarbonate), Cl (chloride), NO₃ (nitrate), SO₄ (sulphate) and K (potassium)) were measured by ion chromatography (DIONEX – ICS - 1,000), which is a specific model of ion chromatography system manufactured by Thermo Fisher Scientific. It is a high-performance liquid chromatography (HPLC) system designed for the separation and analysis of ions in aqueous samples. The samples were first filtered through a 0.45 μm membrane filter. The cations were acidified with 1 per 1,000 nitric acid (HNO₃) surprapur to avoid adsorption and chemical precipitation.

Trace elements (lithium (Li), boron (B), aluminium (Al), vanadium (V), chromium (Cr), manganese (Mn), iron (Fe), cobalt (Co), nickel (Ni), copper (Cu), zinc (Zn), arsenic (As), rubidium (Rb), strontium (Sr), molybdenum (Mo), cadmium (Cd), tin (Sn), caesium (Cs), barium (Ba), lead (Pb), uranium (U)) are all commonly measured using ICP-MS (inductively coupled plasma mass spectrometry). This technique is particularly useful for analysing environmental samples, such as water, soil, and sediment, because it can detect even very low levels of contaminants.

The flows are obtained from two series of differential gaugings with the current meter carried out in the lower Soummam River valley. A current meter is an instrument used to measure the velocity of water in a stream, river or canal. In this case, it is specifically used in the context of gauging the flow of the Soummam River.

In this section, we will analyse the hydraulic relationship between the alluvial aquifer of the lower Soummam and the Oued Soummam. To this end, we will use the results of the measurements made in the section on measuring tools and sample analysis.

Tables 1 and 2 summarise the measurements carried out by differential gauging method. This method expresses the difference of flow between the measuring stations at the level of the Oued Soummam according to the axis of the flow (Sidi Aich, Il-Maten, El-Kseur and Embouchure). This difference is explained either by infiltration and/or sampling in the case of a negative difference, or by drainage in the case of a positive difference.

We can see that there is an increase in the water flow (Q) of the Oued Soummam at Sidi Aich (upstream of the basin). This is explained by the drainage of the alluvial aquifer of the upper Soummam. Furthermore, there is a systematic decrease in the flow between Sidi Aich and El-Kseur, which is explained by the infiltration of the river into the alluvial aquifer, but also by direct withdrawals for irrigation purposes. However, there is an increase in flow between El-Kseur and Embouchure due to the drainage of the alluvial aquifer.

Three types of analyses have been carried out in this section. The first relates to surface water analysis, the second to groundwater analysis and the third to surface and ground water analysis.

During periods of low water (Figure 2), we can see:

• A decrease in surface water concentration along the 0–5 km stretch upstream of Sidi Aich, mainly due to the draining of the upper Soummam aquifer at the level of the Sidi Aich hydrological threshold.

• An increase in concentration in the 5–15.7 km stretch (Remila – Amassine – El-Kseur) which is essentially due to the inflow from the Remila and Amassine tributaries.

• A decrease in concentration is noticeable between the El-Kseur station and the Mellala station over a distance of 15.7–32.4 km and can only be explained by dilution with groundwater. The phenomenon of drainage of the alluvial aquifer is therefore observed.

• A strong increase is observed between the station of Mellala and the embouchure over a distance ranging from 32.4 to 35.8 km, due to the various contributions of the Amizour tributary without the input of the alluvial aquifer.

In contrast, in the high water period (Figure 3), we observe almost the same curve as in the low water period, except that the deconcentration in elements starts just after the Amassine tributary and ends before the Amizour tributary.

The chemical composition of the groundwater is dominated by Na⁺, Ca²⁺, HCO³⁻, Cl⁻ and ⁠, the mineralisation being largely due to the natural chemical dissolution of carbonates from the limestone massif drained by the Remila tributary, Triassic gypsum from the evaporitic formations drained by the Amassine tributary, and anthropogenic activities throughout the basin, such as agricultural fertilisers and domestic and industrial waste.

The water analysed is dominated by Na⁺, Ca²⁺, Cl⁻ and ions in the boreholes in the upstream part during the high-water period (Figure 5). This high level of mineralisation results from the leaching of the Triassic formations, which are carried by the Amassine tributary, then infiltrated in the upstream part (from 0 to 15.7 km), to reach the groundwater. During low water periods, high concentrations of these elements are observed in all the boreholes (upstream and downstream) and in the river water (Figure 4). It should be noted that during this low water period, in the absence of leaching from the Triassic formations, the groundwater loses its hyper-chlorinated to chlorinated facies and the surface water becomes more concentrated in salt due to the lack of contribution from the temporary freshwater tributaries.

The graphical representation of trace elements is dominated by Sr⁺, Li and B ions and follows the same pattern as that of the major elements. They come from the dissolution of evaporitic formations in the waters of the lower Soummam (Figure 5). During the low water period and with the reduction in water flow, leaching is reduced and the teasers in its elements are reduced (Figure 4).

The use of chemical tracers analysed in groundwater and aquifer rocks has made it possible to specify the process governing the increase in water mineralisation. In the upstream part of the basin, from Sidi Aich to El-Kseur, the water is of evaporitic origin, while in the downstream part, from El-Kseur to the mouth, the water is carbonate.

The hydraulic relationship between the river and the alluvial aquifer can also be confirmed by comparing the concentrations of surface water with groundwater in the same upstream or downstream zone (Tables 3 and 4). The upstream sector extends from 0 to 15.7 km and the downstream sector extends from 15.7 to 35.09 km.

In general, it can be concluded that the surface water level of each element in the upstream sector is lower than the groundwater level in the same sector. This increase in groundwater can only be explained by the infiltration of river water into the alluvial aquifer. However, the surface water concentrations of each element in the downstream sector are higher than those in the groundwater in the same sector. This reverse process can only be explained by the river being fed by the alluvial aquifer (drainage) (Tables 3 and 4).

For example, the Cl content of surface water in the upstream part varies from 282.4 to 512.45 mg/L and in groundwater from 481.35 to 2,608.76 mg/L. Surface water contributes to the increase of chloride ions in groundwater through infiltration. However, the Cl content in the surface water in the downstream section varies from 243.27 to 354.74 mg/L and in the groundwater from 46.68 to 594.05 mg/L. In this case, it is the groundwater that contributes to the increase of chloride ions in the surface water, as the river is fed by the alluvial aquifer (drainage) (Tables 3 and 4).

The methodology developed is based on the use of hydrogeochemical tools, which are often a very effective means of studying the phenomenon of exchange between the river and the alluvial aquifer. However, the study by Loukman et al. (2017) of the process of water mineralisation over the entire study area does not provide an understanding of the mineralisation process, nor does it give an idea of the recharge zones or the relationship between the aquifer and surface water (Loukman et al., 2017).

Tables 3 and 4 present the evolution of the content of major and trace elements in surface water and groundwater, respectively (high water period).

The results of the study are of significant relevance to a wide range of fields, particularly in terms of engineering design. The study provides information for engineers and planners in the design of environmental protection systems, such as drainage systems or wastewater treatment systems. In terms of regulations, the study can be useful in drawing up regulations aimed at reducing the risks of pollution and can provide information for political decision-makers in designing more sustainable energy and environmental protection systems. In terms of finance, the study can be useful to investors in their search for sustainable and profitable projects. In terms of ESG (environmental, social and governance), information can be useful to companies and organisations in improving their ESG performance. This valuable information can be used by research, industry and policy makers to improve understanding of the region's hydrological cycle. More specifically, the study provides information on groundwater pollution risks, environmental protection technologies and the potential impacts of climate change in order to develop more efficient, sustainable and resilient water resource management strategies.

Based on variations in concentrations of major and trace elements in surface and groundwater in the Soummam Valley, it was concluded that the river and aquifer are interconnected, and that the river feeds the aquifer upstream and drains it downstream. The analytical sensibility for this study would be to explore the impact of different hypotheses on the conclusions of the study.

Assuming that the concentrations of certain elements in surface water are lower than those measured. This could reduce the vulnerability of the aquifer to pollution for several reasons. Firstly, lower concentrations of pollutants in surface water mean that there are fewer potential pollutants that could seep into the aquifer. Secondly, lower concentrations of pollutants in surface water can lead to a decrease in the flow of water in the river, which could reduce the transport of pollutants to the aquifer.

We can therefore assume that the concentrations of certain elements in the groundwater are lower than those measured. This means that there are fewer potential pollutants that could be used by humans or animals. Also, lower concentrations of pollutants in groundwater can lead to improved water quality, which could have a positive impact on human health and the environment.

It could also be assumed that the concentrations of certain elements in surface water are higher than those measured. Increased concentrations of elements in surface water can lead to an increase in the vulnerability of an aquifer to pollution. This is because pollutants are more likely to enter the aquifer, to be more toxic and to reduce the ability of aquatic organisms to defend themselves against pollution.

It could be assumed that the concentrations of certain elements in the groundwater are higher than those measured. This means that the aquifer is more likely to be contaminated by pollutants. This is because the pollutants are more concentrated and therefore more likely to enter the aquifer.

Based on an analysis of these different hypotheses, it is suggested that the conclusion of interconnection between the river and the Soummam Valley aquifer is robust to variations in pollutant concentrations in surface and groundwater. However, further research is needed to better understand the dynamics of the interconnection, in particular the following factors: the geological conditions of the valley, such as the nature of the rock formations and the depth of the aquifer, climate, which influences rainfall, evaporation and infiltration, and human activity, which can affect water quality and flows.

The results of this study can be used to improve aquifer monitoring. Water managers can focus their monitoring efforts on the most vulnerable areas, develop more effective monitoring protocols and improve the interpretation of monitoring data.

The results of this study can be used to develop more effective monitoring protocols. By knowing which types of contaminants are most likely to pollute the aquifer, water managers can choose which parameters to monitor and the appropriate monitoring frequencies. This study shows that concentrations of major elements such as sodium, calcium, bicarbonate, chloride and sulphate and trace elements, such as strontium, lithium and boron in surface water are an important factor in the vulnerability of the aquifer. The results of this study can also be used to improve the interpretation of monitoring data.

By understanding the factors that can affect contaminant concentrations in groundwater, water managers can better understand the implications of monitoring data. This study shows that concentrations are dominated by major elements such as sodium, calcium, bicarbonate, chloride and sulphate in the upstream boreholes during the high water period. This high level of mineralisation results from the leaching of Triassic formations, which are transported by the Amassine tributary and then infiltrated in the upstream part to reach the groundwater. Trace elements such as strontium, lithium and boron in the groundwater are a major factor in the vulnerability of the aquifer.

Water managers could focus their monitoring efforts on areas where high concentrations of major elements, such as the natural chemical dissolution of carbonates from the limestone massif drained by the Amassine tributary, as well as anthropogenic activities throughout the basin, such as agricultural fertilisers, domestic and industrial discharges, etc., are present.

By incorporating the results of this study into their monitoring programme, water managers can better understand the vulnerability of the aquifer to pollution and take steps to protect this precious resource. In terms of monitoring protocols, the study could recommend that concentrations of major elements be monitored every six months in the most vulnerable areas, and every year in other areas. Trace element concentrations should be monitored every 12 months in all areas.

For the interpretation of monitoring data, the study could recommend that water managers take action if contaminant concentrations exceed the following thresholds:

Major elements: 200 mg/L for sodium, 20 mg/L for calcium, 600 mg/L for bicarbonate, 200–500 mg/L for chloride and 400 mg/L for sulphate.

Trace elements: 50 μg/L for strontium, 100 μg/L for lithium and 1 mg/L for boron. These recommendations would provide water managers with clear guidelines for improving aquifer monitoring.

This section discusses the limitations of the methodology in further detail:

Data scope and sampling frequency:

• Spatial coverage: The study's findings are limited to the specific locations of boreholes, wells and stations. It might not capture pollution dynamics in areas lacking monitoring points.

• Temporal resolution: Sampling during high and low water periods provides valuable insights, but it might miss short-term variations or pollution events occurring between those periods.

• Sampling frequency: Infrequent sampling could potentially miss transient pollution events or seasonal variations, potentially leading to an incomplete understanding of water quality dynamics.

Monitoring system sustainability:

• Long-term maintenance: Ensuring the continuity and quality of data collection over extended periods can be challenging, involving ongoing costs, personnel, and equipment upkeep.

• Data management and analysis: Effective utilisation of the monitoring system's data requires robust data management and analysis capabilities, which might not always be available.

Addressing uncertainty and variability:

• Natural fluctuations: Natural factors like rainfall and groundwater flow can introduce variability in water quality data, making it challenging to isolate pollution signals.

• Data uncertainty: Measurement errors and uncertainties associated with sampling and analytical techniques can influence the accuracy of results.

External factors influencing water quality:

• Costs and resources: Implementing and maintaining a comprehensive monitoring system can be resource-intensive, potentially limiting its feasibility in certain contexts.

• External influences: The methodology might not fully account for external factors like land use changes, upstream pollution sources or climate change impacts that could affect water quality dynamics.

To build upon this study's valuable insights and overcome its limitations, future research could explore exciting avenues:

• Denser monitoring networks: Expanding beyond isolated boreholes to include wider coverage with advanced sensors capturing real-time data would reveal the hidden dynamics of river–aquifer interactions.

• High-resolution sampling: Employing continuous or more frequent sampling strategies would capture fleeting pollution events and unveil the nuanced seasonal dance of water quality fluctuations.

• Advanced modelling and analysis: Utilising sophisticated data processing and modelling techniques could disentangle natural variability from pollution signals, providing a clearer picture of cause-and-effect relationships.

• Cost-effective solutions: Investigating innovative monitoring technologies and data management strategies could make comprehensive water quality assessments more accessible and sustainable in resource-constrained contexts.

• Holistic assessments: Expanding the scope beyond the river–aquifer interface to include upstream pollution sources, land use changes, and climate influences would paint a complete picture of the factors shaping water quality dynamics.

By pursuing these research frontiers, we can overcome the limitations of this study and gain a deeper understanding of the intricate dance between river and aquifer, ultimately safeguarding the precious water resources of the Soummam Valley and beyond.

This study presents the results of the differential gauging and hydrogeochemical tracing carried out during two sampling campaigns during high and low water periods in the Oued Soummam River and in the alluvial aquifer of the lower Soummam Valley.

The application of differential gauging in the case of the low water period made it possible to observe the existence of an increase in the water flow of the Oued Soummam at Sidi Aich (upstream of the basin) whose strong flows are recorded in the upstream and downstream stations of the basin and the weakest at the level of the intermediate station of El-Kseur.

Hydrogeochemical tracing applied to the surface water samples revealed a decrease in the concentration of major and trace elements in the water in the stretch from 0 to 5 km (Sidi Aich – Remila) and an increase in concentration in the stretch from 5 to 15.7 km (Remila – Amassine – El-Kseur). However, a decrease in the content of chemical elements in the surface water was marked due to the feeding of the river by the alluvial aquifer in the downstream part along the stretch from El-Kseur to the Embouchure.

The upstream sector (Sidi Aich – El-Kseur) has a high capacity for water infiltration into the aquifer. For this reason, it is advisable to avoid setting up household waste dumps, drilling boreholes or wells and discharging raw sewage without treatment.

The mineralisation of the water has increased due to the leaching of the evaporite formations, and the water is sodic-chlorinated and has a high salt content. Thus, the hydraulic relationship between the river and the alluvial aquifer is highlighted through the feeding of the alluvial aquifer by the river at the upstream end and drainage at the downstream end.

This study has therefore highlighted the complexity of the aquifer–river functioning in the Soummam Valley in the absence of a continuous flow measurement station. In particular, it has highlighted the predominance of the geological and hydrogeological environment, which favours the recharging of the alluvial aquifer by the tributaries of the slopes containing Triassic formations of evaporitic type on the one hand, and a limestone slope of Miocene age on the other, but which can also act as a brake on the exploitation of the water resource by increasing the hardness of the water.

The installation of continuous flow measurement stations along the watercourse could be an interesting additional tool to quantify flow variations and contaminant flows, as well as the different dilution ratios according to the chemical element content of groundwater and surface water.

The results of the study can be used to develop pollution-vulnerable zone maps to identify areas where groundwater is most likely to be contaminated by surface water, to develop simulation models of the river–alluvial aquifer interconnection and to develop water resource management strategies in the lower Soummam Valley. The results of the study are therefore of vital importance for the protection of groundwater in the region and worldwide.

The findings of the study are relevant to a wide range of stakeholders, including engineers, planners, policy makers, investors and businesses. They provide valuable information on the risks of groundwater pollution, environmental protection technologies and the potential impacts of climate change. This information can be used to develop regulations to reduce pollution, develop more efficient, sustainable and resilient water resource management strategies, and identify sustainable and cost-effective projects.

We would like to thank the ‘Laboratory of Applied Hydraulic and Environmental Research’ and the ‘Scientific and Technical Research Centre of Physical and Chemical Analysis’ of the University of Bejaia, where we carried out the manipulations, for their support and help, as well as everyone else who helped us to carry out the experiments successfully and to make the best use of the results.

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

The authors declare there is no conflict.