Evaluating the quality and nutritional content of bottled waters in Algeria

Water is vital for all living species, essential for the functioning of the human body and performs construction, thermoregulatory, transport and chemical roles (Imbert 2016). It is available for human consumption in various forms, including in bottles, and is provided from several sources, such as springs and aquifers (Tatou et al. 2017; Tashakor & Modabberi 2020; Lee et al. 2021; Najafzadeh et al. 2021; Turhan et al. 2021).

Two different types of Algerian bottled waters are to be distinguished: the natural mineral waters which are underground waters of stable physicochemical composition, exempt from all forms of contamination. They are subjected to legal protection measures, in particular for the perimeter of the catchment area of the water sources. Whereas the spring waters are those of exclusively underground origin, suitable for human consumption, microbiologically safe and protected against the risks of contamination (Jora 2004).

To characterize and prevent any potential health effects, the physicochemical parameters (pH, dry residue, anions, cations, and certain trace elements) of bottled waters in Algeria are regularly assessed at their origin. Additionally, they must be devoid of parasites, pathogenic microorganisms, and any fecal contaminants (Jora 2006, 2015).

Worldwide, numerous studies on bottled waters have shown differences in physicochemical composition and quality that the elements that contain can be benefic or may be a health burden when they are present in concentrations out of certain limits (Galan et al. 2002; Güler et al. 2002; Albertini & Dachà 2007; Sengupta 2013; ĆUK et al. 2016; Imbert 2016; Quattrini et al. 2017; Balejčíková et al. 2020; Crespo et al. 2021; Ankon et al. 2022, Tapias et al. 2022). Whereas, in Algeria fewer investigations have been performed to assess the quality of bottled waters (Hazzab 2011; Hazzab 2012; Sekiou & Khellil 2014; Labadi & Hammache 2016; Sekiou & Tamrabet 2022), few to evaluate the incidence on human health (Bencheikh et al. 2021; Kerdoun et al. 2021; Bouteldjaoui & Taupin 2023) and none to estimate the potential contribution of bottled water minerals to dietary intake.

In fact, Hazzab (2011, 2012) examined the evolution in legislation with respect to the exploitation, manufacturing and marketing of bottled waters, while Sekiou & Kellil (2014) and Sekiou & Tamrabet (2022) discussed the classification of bottled mineral and spring waters in Algeria and the Maghreb regions using graphical and multivariate approaches. Their research revealed that bottled waters can be divided into two and four main groups, respectively. Labadi & Hammache (2016) conducted a comparison between mineral and spring waters produced in Algeria.

The work of Bencheikh et al. (2021), on 42 Algerian bottled water brands, concluded that only 97.62% of samples were compliant with the Algerian standards and hence were safe for all categories of human consumption; whilst Kerdoun et al. (2021) investigated fluoride levels in bottled waters in southern Algeria, finding a maximum value of 1.65 mg/l.

In Algeria, there has been no prior study conducted to assess the potential contribution of the mineral composition of bottled waters to Dietary Reference Intakes (DRIs). However, abroad, the DRI system of bottled waters has been investigated in many research works (ĆUK et al. 2016; Gazan et al. 2016; Sajjala et al. 2019; Wysowska et al. 2022).

Gazan et al. (2016) indicated that adequate consumption, particularly of natural mineral water, is linked to better diet quality. Additionally, research conducted in the French population indicates that higher water intake is associated with increased consumption of essential micronutrients. In their approach to estimate the quality of bottled waters in Serbia, ĆUK et al. (2016) noticed that DRI contribution of Na⁺ (11.76–82.73%) and Cl⁻ (5.2–38.1%) where the lower value refers to children 1–3 years of age and the upper value refers to lactating females. Simultaneously, the magnesium intake related to limestone/dolomitic ranges from 18.2% for girls (14–18 years of age) to 65.66% for children (1–3 years of age), while calcium absorption is highest for the latter age group. In a comparison of local-Omani and imported-UAE bottled waters, Sajjala et al. (2019) concluded that imported mineral water brands fulfill the DRI requirements better than drinking water brands sold in Oman; while on the dietary sodium contribution scale, Omani bottled water brands are categorised as very low-sodium and UAE brands are categorized as sodium-free. Wysowska et al. (2022) found that the doses of aluminum, chromium and nickel calculated for mineral water were 1 to 2 times higher than the RDI for children and 1.5 to 7 times higher than the RDI for adults; whereas the average concentration of arsenic in bottled waters provided 9% RDI for children and 27% for adults, chromium: 37% RDI and nickel: 46% for adults.

Using different empirical, graphical and multivariate techniques, the study discusses the water quality and classification according to the Algerian (Jora 2004, 2006, 2015), the European (EU 1998, 2003, 2009), and the World Health Organization (WHO 2005, 2008) legislations.

Several methods for characterizing and classifying groundwater bottled waters have been presented in the literature (Tanaskovic et al. 2014; Bodrud-Doza et al. 2016; Alfaifi 2019; Mfonka et al. 2021; Taşan et al. 2022). They employ various multivariate statistical methods such as hierarchical cluster analysis (HCA), principal component analysis (PCA), and empirical and graphical tools as representation aids in comprehending complex data for improved insight into the quality and classification of bottled waters (Alfaifi 2019; Everest & Özcan 2019; Mfonka et al. 2021). This combination assists in identifying the hydrochemical mechanisms that influence the chemical composition of groundwater (Alfaifi 2019).

DRIs are universally accepted benchmarks for nutrient values. They encompass the Estimated Average Requirement (EAR), Recommended Dietary Allowance (RDA), Adequate Intake (AI), and Tolerable Upper Intake Level (UL). By considering the daily consumption of bottled water per age and gender, the nutritional contribution of bottled water to the diet was determined. This involved multiplying the concentration of nutrients by the daily intake and comparing it to the RDA or UL (NLM 2000).

The bottle labels used to obtain the physical and chemical characteristics of the waters were subjected to verification with the water producers and their association. The physicochemical parameters mentioned on labels are: pH, fixed residue and silicate (SiO₂), cations such as calcium (Ca²⁺), magnesium (Mg²⁺), sodium (Na⁺), potassium (K⁺) and anions such as bicarbonate (HCO₃⁻), sulfate (SO₄²⁻), chloride (Cl⁻), nitrate (NO₃⁻), nitrite (NO₂⁻) and fluoride (F⁻).

The initial assumption is that most consumers across Algeria perceive bottled mineral and spring waters as alike in terms of their essential components. Consequently, the objective of this article is to update the inventory and the geographic distribution of production units, to investigate the quality, compliance with national and international standards, classification and to estimate the potential contribution of dietary nutrients of the Algerian bottled mineral and spring water brands.

From a hydrogeological perspective, Algeria is subdivided into two domains (Northern and Southern). In the Northern Domain, the complex tectonic history has segmented the major geological units from the Mesozoic to the Cenozoic, resulting in a large number of relatively small and spatially limited compartmentalized aquifer units (Figure 2).

Within this domain, the main types of aquifers are:

• a.

  Recent Cenozoic and Quaternary unconsolidated sedimentary aquifers in the Coastal Plain

• b.

  Mesozoic-Cenozoic sandstone and limestone aquifers in the mountainous areas (Atlas 2023).

In this study, 63 bottled water brands chosen for this study represent the current highest selling and most consumed brands on a daily basis by the Algerian population. Thirty-three and 30 brands of bottled spring and natural mineral waters, respectively, available on the national market were purchased in randomly selected local supermarkets (Table 1).

The spatial distribution of bottled water production sites can be influenced by several factors, and it can vary widely across different regions and countries. Aspects that determine the spatial distribution of these sites might be: water source availability, infrastructure and logistics, market proximity and economic factors. In the case of Algeria, the production units are located in the vicinity of the water sources and near transportation hubs to facilitate efficient distribution.

Algeria is a vast country, made up hydrologically of five hydrographical basins (AHS, CZ, CSM, OCC and Sahara) and administratively of 58 provinces. On the national market, bottled waters produced and sold locally fall exclusively into mineral water and spring water; the most widespread (63 brands) are spatially distributed by hydrographical basin and province (Table 1).

Among the hydrographical basins, four (AHS, CZ, CSM and OCC) border the southern coast of the Mediterranean Sea, which hold 70 and 85% of the total bottled mineral and spring waters sites, respectively. Within this geographical space, the Algerois–Hodna–Soumam (AHS) holds 30% of the total bottled mineral water, ahead of Constantinois–Seybous–Mellegue (CSM) with (23.3%), Oranie–Chot–Chergui (OCC) with (16.7%) and Chelif–Zahrez (CZ) with (0%). Meanwhile, bottled spring water sites are located within the AHS (51.5%), CSM (24%), OCC (6.1%) and CZ (3.1%).

In this regard, the provinces of Béjaia and Blida (AHS) each contribute largely to 10% of the mineral water production units among the northern hydrographical basins (AHS, CZ, CSM, and OCC); while 27.3 and 12.1% of the spring water production units for Béjaia and Bordj–Bou–Arreridj, respectively, are dominant (Table 1).

It is clear that the province of Béjaia and the (AHS) hydrographical basin have the most bottled water producing facilities in the country.

However, the Sahara hydrographical basin hosts 30 and 15% of total bottled mineral and spring waters sites, respectively, based mostly north of the latitude 34°00′00″N (Figure 2). In this space, the province of Biskra represents the main production pole of the region. In fact, the bottled mineral water production units are distributed throughout the provinces of Biskra (13.3%), Tebessa (10%), Laghouat (3.3%) and Ghardaia (3.3%); meanwhile, the bottled spring water production units are less significant, where the provinces of Biskra, Adrar, Laghouat and Ain-Ghezzam contribute with 6, 3, 3 and 3%, respectively (Table 1).

The bottle labels used to obtain the physical and chemical characteristics of the waters were subjected to verification with the water producers and their association. The physicochemical parameters mentioned on labels were: pH, fixed residue, calcium (Ca²⁺), magnesium (Mg²⁺), sodium (Na⁺), potassium (K⁺), bicarbonate (HCO₃⁻), sulfate (SO₄²⁻), chloride (Cl⁻), nitrate (NO₃⁻), nitrite (NO₂⁻), fluoride (F⁻) and silicate (SiO₂); in addition to total dissolved solids (TDS) and the total hardness (TH) which were calculated (Simler 2023). The water quality data for the major ionic species were tested for validation and reliability prior to further use.

Since there is a insufficient data about water fluoride (F⁻) and silicates (SiO₂), they won't be considered in this study. In this context, only bottled waters containing fluoride levels exceeding 1 mg/l must include this information on their labels (Jora 2006).

Given that most of the sites for bottled mineral and spring waters (96.70 and 94%, respectively) are located north of the latitude 34°00′00″N (Figure 2), the investigation will thus focus exclusively on this part of the country.

Graphic tools were used to determine the major factors influencing the groundwater systems. Indeed, Chadha diagram (Chadha 1999), a modified version of the Piper diagram and the expanded Durov diagram, is a proposed new diagram for geochemical classification of natural waters and interpretation of chemical data.

PCA and HCA are two useful multivariate techniques to understand the sources and influencing factors of groundwater chemistry, and have been used in many groundwater studies (Abdi & Williams 2010; Sekiou & Kellil 2014; Sekiou & Tamrabet 2022).

The DRIs are a set of reference values for vitamins, minerals, and other nutrients important to human health established by the Institute of Medicine of the U.S. National Academy of Sciences (NLM 2000). The DRI system includes the RDA, AI and tolerable UL, and it is used to calculate contributions of the elements essential for human health according to age group and gender (Suitor & Murphy 2013; Ćuk et al. 2016).

An RDA is an average daily dietary intake level, sufficient to meet the nutrient requirements of nearly all (97–98%) healthy individuals in a group.

If sufficient scientific evidence is not available to establish an RDA, an AI is usually developed. UL is the highest level of daily nutrient intake that is likely to pose no risk of adverse health effects to almost all individuals in the general population (Suitor & Murphy 2013).

Diagrammes (Simler 2023), Xlstat (Microsoft, 2016) and QGIS (version 3.10, QGIS 2019) softwares were used for water hydrochemistry, statistical analyses and spatial analysis, respectively.

The physicochemical parameters mentioned on bottled mineral and spring water labels are listed in Table 2. Generally, it was observed that the concentrations and distributions of hydrochemical parameters of the mineral water showed significant spatial variations. The pH values ranged from 6.50 to 7.90, with an average of 7.28 for mineral water, and from 6.60 to 7.80, with an average of 7.28. As bottled waters originated from groundwater, it is obvious that most of the water pH tended to be alkaline (Raja et al. 2021); nonetheless few were slightly acidic because of their carbonated nature (Mouzaia (#38), Ben-Haroun (#40)). These waters are found in areas of large tectonic faults where their richness in CO₂ is associated with different regional geological-structural features and related to granite intrusions and volcanic rocks (Marinković et al. 2013). The water saturated with CO₂ intensifies the process of carbonate weathering and easily dissolves the carbonate minerals available in its flow path. This process has increased the soluble ion content like chloride, sodium, potassium, magnesium, and bicarbonate ions in the groundwater (Hwang et al. 2017). The pH of the bottled waters was within the allowable limits (Jora 2006; WHO 2008; Jora 2015) (Table 2). In mineral water, the variation ranges of (TDS), (FR) and (TH) values were significant (33.00–4,492.00 mg/l), (180.00–2,800.00 mg/l) and (196.94–3,907.77 mg/l) with higher mean values (867.47 mg/l), (589.57 mg/l) and (681.20 mg/l) than in spring water with variation range of (127.00–1,520.00 mg/l), (100.00–953.00 mg/l) and (71.35–1,039.40 mg/l) and mean values of 768.5, 501.8 and 580.8 mg/l, respectively (Table 2). The TH of the bottled waters is mostly attributed to the salts of Ca⁺² and Mg⁺² where temporary hardness here is more important than permanent hardness, as the concentration of HCO₃⁻ is comparatively more than those of Cl⁻ and SO₄⁻².

For TDS and TH, no guidelines were established for mineral water, however, some of the spring water values (Brands #4 or 3% for TH and #21 or 3% for TDS) exceeded the recommended guidelines for drinking spring water (WHO 2011; Jora 2015) (Table 2).

TDS is a parameter that counts all dissolved minerals in the water; out of them Ca⁺² and Mg⁺² constitute mainly the TH and are beneficial for human health. In fact, hard drinking water generally contributes a small amount toward the total calcium and magnesium needed in the human diet (Galan et al. 2002; Sengupta 2013). Moreover, according to Galan et al. (2002) and WHO (2008), neither TDS nor TH has known adverse health effects.

However, the dissolved salts in hard water, particularly calcium and magnesium at high levels, may have some adverse health consequences. Despite that, individuals are mostly protected against excessive calcium intakes due a strictly controlled intestinal absorption system by means of 1,25-dihydroxyvitamin D, the hormonally active form of vitamin D (Sengupta 2013).

Within the intestines, calcium can interact with iron, zinc, magnesium, and phosphorus, limiting their absorption. On the other hand, renal insufficiency, which is linked to a considerably reduced ability to eliminate magnesium, is the main cause of hypermagnesemia (WHO 2008; Sengupta 2013). Increased use of magnesium may result in diarrhea. The latter may happen when high concentrations of both magnesium and sulfate (250 mg/l) are present in bottled waters (Fiorentini et al. 2021). In addition, magnesium supplements, as opposed to magnesium in the food, have been linked to laxative effects when used in excess (Fiorentini et al. 2021).

In spring water, the NO₃⁻ and NO₂⁻ concentrations range from 0.73 to 49.00 mg/l and 0.00 to 0.06 mg/l with mean values of 12.926 and 0.014 mg/l, respectively. At the same time, the NO₃⁻ and NO₂⁻ values for mineral water varied between 0.00–21.80 mg/l and 0.00–0.50 mg/l, with mean values of 6.951 and 0.091 mg/l, respectively.

Concentrations of NO₃⁻ ions are within the permissible limits for both drinking mineral and spring waters (EU 2003; Jora 2006; Jora 2015). Meanwhile, the mineral water contains more NO₂⁻ ionic concentration as compared to the spring water; moreover, it exceeds the maximum permissible limit for drinking some brands (brand #63 or 3%) (EU 2003; Jora 2006) (Table 2).

The known toxic effects of nitrate exposure result from the conversion of nitrate to nitrite (Balejčíková et al. 2020). In fact, consuming a water rich in nitrate can cause infant methemoglobinemia particularly for babies under six months old (MN-US 2023). Methemoglobinemia can result in serious illness or death. Risk of specific cancers and birth defects may be increased when nitrate is ingested under conditions that increase formation of N-nitroso compounds (Ward et al. 2018).

With the exception of the mineral water which is exempt from allowable limits, for the rest of the physicochemical parameters of bottled waters, their concentrations are within the allowable limits and consequently do not pose any serious health risks.

Empirical and graphical methods have the advantage of providing valuable information on the presence and chemical facies of groups (Cüneyt et al. 2002).

According to EU Directive 2009/54 EEC (2009), (Table 3), nearly 25% of the Algerian bottled waters belong to the category of ‘suitable for low sodium diets, or Na⁺ < 20 mg/l’ which is recommended for individuals with hypertension. Consuming too much salt has long been linked to a number of cardiac and circulation issues (Rosinger et al. 2021). Only one mineral water brand (Ben-Haroun (#40)) had Na⁺ > 200 mg/l which is suggested for intense physical activities (Quattrini et al. 2017).

The major ion concentrations were lower in the spring water than in the mineral water. The relative abundances of major cations and anions were ranked in the order Ca⁺² > Na⁺ > Mg⁺² > K⁺ and HCO₃⁻ > SO₄⁻² > Cl⁻ > NO₃⁻ > NO₂⁻, respectively, for both waters. (Ca⁺² + Na⁺) together accounted for 82% of the total cations, whereas HCO₃⁻ + SO₄⁻² accounted for 85% of the total anions for both the spring and mineral waters.

Bottled waters were primarily used for drinking. The TDS classification of the bottled waters with respect to EU (2009), (Table 3) revealed that all the spring water and 97% of the mineral water belonged to the low/intermediate mineral content category (50 mg/l < TDS < 1,500 mg/l), while only 3% of the mineral water brands were rich in mineral content (TDS > 1,500 mg/).

The Ca-Mg-HCO₃ type could be attributed to weathering processes which lead to dissolution of minerals from parent rock and weathering of carbonates and dissolution of silicate minerals (Gaikwad et al. 2020), while the Na + K-Cl and Na + K-HCO₃ types could be explained by ion exchange mechanisms between groundwater and subsurface geologic strata (limestone) (Ueda & Kusakabe 2015).

A correlation matrix serves to investigate the interactions between water quality indicators and the origin of water-soluble compounds in the location of water production (Kothari et al. 2021).

A Pearson correlation matrix was created using the actual values of the 13 variables chosen for statistical analysis (Table 4), including pH, FR, TDS, TH, Ca²⁺, Mg²⁺, Na⁺, K⁺, Cl⁻, HCO₃⁻, SO₄²⁻, NO₃⁻ and NO₂⁻. The effects of various geological and human activities on water quality may result in high or low correlations between hydrochemical parameters (Nong et al. 2019).

The correlations between the ions were calculated (Table 4) and showed strong (r> 0.80) to moderate (r> 0.60) positive correlations with significant P<0.01 obtained for Ca⁺²-SO₄⁻², Ca⁺²-HCO₃⁻, Na⁺-SO₄⁻², Na⁺-Cl⁻, Cl⁻-HCO₃⁻, Ca⁺²-Mg⁺², Mg⁺²-SO₄⁻², Mg⁺²-HCO₃⁻, SO₄⁻²-Cl⁻, SO₄⁻²-HCO₃⁻, for the spring water on one hand, and strong to moderate positive correlations with significant P<0.01 for Ca⁺²-Mg⁺², Ca⁺²-Na⁺, Ca⁺²-SO₄⁻², Ca⁺²-Cl⁻, Ca⁺²-HCO₃⁻, Mg⁺²–Na⁺, Mg⁺²-Cl⁻, Mg⁺²-HCO₃⁻, Na⁺-SO₄⁻², Na⁺-Cl⁻, Na⁺-HCO₃⁻, SO₄⁻²-Cl⁻, SO₄⁻²-HCO₃⁻, Cl⁻-HCO₃⁻ for the mineral water on the other hand.

(TDS), (FR) and (TH) were highly positive inter-correlated (r> 0.90). With the exception of pH, K⁺, NO₃⁻ and NO₂⁻, they were highly positive correlated (r> 0.90) with (Ca²⁺, Na⁺, HCO₃⁻) and strongly positive correlated (r> 0.80) with ions (Na⁺, Cl⁻ and SO₄²⁻) of both bottled spring and mineral waters.

The high positive correlation between (TDS), (FR) and (TH) showed that they are interrelated, while their correlations with the major ions provide information on the main parameters controlling mineralization of the bottled waters. In the mean time, the high correlation between TH and Ca²⁺ indicates that the TH is of calcic type. The high positive correlation (r> 0.9) between Ca²⁺ and HCO₃⁻ could be attributed to weathering processes which lead to dissolution of minerals from parent rock and weathering of carbonates and dissolution of silicate minerals (Gaikwad et al. 2020), meanwhile the significant positive correlation (r> 0.7) between Na⁺ and HCO₃⁻ could be explained by ion exchange mechanisms between groundwater and subsurface geologic strata (limestone) (Ueda & Kusakabe 2015).

The application of the technique of PCA resulted in the extraction of several components, whose first three explained 75.31% of the total variance (Table 5) which can be considered sufficient for analyzing the approach of variables and individuals in order to identify the main sources of hydrochemical variation (Ghodbane et al. 2022). It was verified that the first (F1), the second (F2) and the third (F3) components explained, respectively, 58.37, 9.23 and 75.31% of the total variance.

Besides, Table 5 shows the weights of the variables that most contributed to these principal components. The parameters that most contributed positively to F1 were Ca⁺², Na⁺, Cl⁻, SO₄⁻², HCO₃⁻, FR, TH and TDS; positively to F2 were NO₂⁻ and K⁺, while NO₃⁻ contributed negatively to it. The only parameter that contributed positively to F3 was pH.

The component F2 indicates the increase in NO₂⁻ concentration, probably leached from soil at recharge spaces and by conversion of the existing NO₃⁻ through denitrification (NO₂⁻, NO₃⁻). This explains the fact that the extracted variables are inversely correlated. The concentrations of NO₂⁻ and NO₃⁻ in bottled waters were low which is the case of the natural nitrate concentration in groundwater under aerobic conditions that depends strongly on soil type and on the geological situation (Khan et al. 2021). However, monitoring the bottled waters is essential owing to potential health hazards such as cancer development, congenital birth defects, cardiovascular disease, high blood pressure, and effects on the neurological system (Gibson et al. 1987).

Two groups of parameters correlating significantly with F1 had an impact on the classification of bottled waters along F1 (Figure 5). In fact, for the first group, the brand water (#40), located on the extreme positive side of the F1 axis, had TDS = 3,907.77 mg/l or (>1,500 mg/l) and TH = 1,291.52 mg/l, which could be classified, based on EU Directive 2009/54 EEC (2009), as rich mineral-very hard water; followed in descending order of TDS values, or toward the left side direction, by a cloud of water brands (37, 38, 58, 21 till 19, 20, 32) with 50 mg/l < TDS < 1,500 mg/l, which form the second group and could be classified in the low to intermediate range of mineral content waters (Figure 5). This group embraced the majority of water brands (98%) and had a large span of TH values (32.3–647.50 mg/l) which covered all the categories of waters with respect to (soft ≤ TH ≤ very hard) water.

The first main cluster (G1) contained the majority of the water brands (62-brands or 98%), which all ranged in (71.35 mg/l ≤ TDS ≤ 1,285.00 mg/l, median = 546.50 mg/l) belonging to the range (50 mg/l < TDS < 1,500 mg/l) (EU 2009). G1 is divided into two major sub-clusters subsequently; they could be classified in the range of low mineral content waters (50 mg/l < TDS < 500 mg/l) for the first (sub-G1₁, left) and intermediate mineral content waters (500 mg/l < TDS < 1,500 mg/l) for the second (sub-G1₂, right). Both sub-clusters were composed of 100 and 97% of spring and mineral water brands, respectively.

Further common features for this cluster of waters were HCO₃^–'s range (24.4 mg/l ≤ HCO₃^– ≤ 671.00 mg/l, median = 263.00 mg/l) within which only the brands 37, 38 and 58 were bicarbonate (>600 mg/l); Na⁺ was ranging between (2.00 mg/l ≤ Na⁺ ≤ 185.00 mg/l, median = 23.00 mg/l), were non-sodic (Na⁺ < 200 mg/l) and 15 brands were suitable for low sodium diet (brands 16, 19, 22, 23, 26, 32, 34, 41, 44, 48, 49, 53, 59, 60, 62); Ca²⁺ was ranging between (4.60 mg/l ≤ Na⁺ ≤ 143.00 mg/l, median = 66.50 mg/l) where all brands were (Ca²⁺ < 150 mg/l); with respect to Cl⁻, only brand 21 was less than 200 mg/l, SO₄²⁻ was less than 200 mg/l except for the spring water brands 31, 4 and 17; whereas Mg²⁺ was below the level of 50 mg/l except for the mineral water brands 38, 37 and 58.

The second main cluster (G2) embraced only one mineral brand (#40) with extreme values of TDS = 3,907.77 mg/l, HCO₃^– = 1,809.30 mg/l, Na⁺ = 680.00 mg/l and Ca²⁺ = 413.00 mg/l. So, it is highly rich in mineral salts, very hard and bicarbonate. Consequently, it could be recommended for therapeutic purposes.

In general, the clustering results were based on the water mineralization, which resulted in two classes in the following order of mineralization: G2 > sub-G1₂> sub-G1₁, where G1 = sub-G1₂ + sub-G1₁.

As a result, the two multivariate tools (PCA and HCA) used in this case were quite consistent.

Additionally, the association between the chemical compositional characteristics and mineral water brands was well illustrated using the HCA dendrogram and PCA plot.

According to this study's findings, TDS and TH are the most crucial factors in classifying and separating the investigated water brands. In their work, Güler et al. (2002) stated that TDS is the most discriminating characteristic in the categorization of water brands.

In all groups taken into consideration, the key macrocomponent elements having significant contributions to DRI which exceeded the tolerable UL are: Ca for Ben-Haroun Brand (#40) with respect to Adults_Men >19 years (123.90%) and Females_Breastfeeding 19–51 years (128.03%); and Na for the same brand with respect to the categories of Boys_14–18 years (117%), Adults_Men > 19 years (148.36%), Adults_Women >19 years (108.80%), Females_Pregnant 19–51yr (104.27%) and Females_Breastfeeding 19–51 years (140.53%).

Other essential macrocomponent elements (Mg and Cl) have contributions to DRI up to 84.38 and 57.04% in the groups of children_1–3 years and adults_Men >19 years, respectively.

The daily intake of K is the least available in the mineral water and does not cross the threshold of 1.34% for all categories.

Generally, the Algerian bottled waters contributed substantially to the daily intake for major mineral elements (Ca, Na, Mg, K and Cl) and even exceeded the tolerable UL for some macroelements (Ca and Na).

Many studies (WHO 2005) reported that water may provide up to 20% of the required total daily intake of major minerals; whereas for the majority of other elements, drinking water provides <5% of total intake. However, in their study on bottled water quality in Serbia, Ćuk et al. (2016) stated that some Serbian waters contributed significantly to the DRI for Ca (28.8% for children and 44.64% for adults), Mg (65.66 and 58.36%), Na (147 and 254.2%), and Cl (32 and 38.13%). Whereas, Reddy & Al-Dawery (2018) mentioned that, in Oman's local and imported bottled waters, local and imported brands contribute to DRI about 4.4 and 18.8% for adults' Ca, respectively. Similar results in a Bangladesh study (Ahmed et al. 2016) reported that the local bottled brands contribute a maximum of 3.3% to DRI for adults' Ca. Meanwhile, as stated by the previous study, the maximum magnesium contribution by local brands was 7.3% to DRI for adults.

Several epidemiological studies (Azoulay et al. 2001; Nguyen et al. 2010; Sasikaran et al. 2012; Sunday & Henrietta 2015) have indicated that the consumption of water could serve as a crucial source of mineral intake, given that minerals present in water are in ionic form and readily absorbed by the gastrointestinal system. In fact, minerals such as Ca²⁺, Mg²⁺, and Na⁺ in water play significant physiological roles, and insufficient intake of these minerals may increase the risk of various diseases.

Throughout all stages of life, the intake of Ca²⁺ is essential, with particularly high demands during childhood, fetal growth, pregnancy, and breastfeeding (Azoulay et al. 2001; Sunday & Henrietta 2015). Epidemiological and clinical evidence suggests a negative association between calcium intake and the incidence of osteoporosis. Especially for adult females, a calcium-rich diet may reduce the risk of age-related bone loss and hip fractures (Azoulay et al. 2001; Sunday & Henrietta 2015).

Epidemiological research (Azoulay et al. 2001; Sasikaran et al. 2012) has revealed a negative correlation between Mg²⁺ intake and the risk of sudden death, cardiac arrhythmias, and ischemic heart disease. Moreover, studies indicate a negative relationship between magnesium levels in water and the prevalence of heart diseases. Notably, magnesium from water is highly bioavailable, being absorbed approximately 38% more efficiently than magnesium from food (Azoulay et al. 2001), suggesting that opting for water rich in Mg²⁺ could be advantageous.

In modern society, Na⁺ consumption tends to exceed recommended limits, contrasting with lower intakes of Ca²⁺ and Mg²⁺. Several studies (Azoulay et al. 2001; Grillo et al. 2019) have highlighted the association between high Na⁺ intake and hypertension development. These findings underscore the potential benefits of restricting dietary sodium, particularly for individuals on a low-sodium diet, as certain water sources may inadvertently contribute to elevated sodium intake levels that could pose health risks (Azoulay et al. 2001; Sasikaran et al. 2012; Grillo et al. 2019).

Despite the fact that they cannot replace the minerals that people acquire via their diet, the minerals in water can be a substantial complement. The current study provides an updated survey of the mineral and spring waters produced and marketed in Algeria, as well as information on their spatial national distribution and comparative quality assessment. It also compares these waters to national and international standards, draws up the implications as they relate to public health, and estimates the contribution of minerals to dietary intake. A dataset of 63 bottled water brands comprised of 30 mineral and 33 spring water brands, provided by local supermarkets, was used for the investigation.

A number of conclusions may be drawn from this study, including:

• •

  Nationwide, the AHS hydrographical basin seems richer in spring water units (51.5%), but the mineral water units dominate the AHS and the Sahara hydrographical basins (30% each); provincially, Bejaia-(AHS) leads in spring water production units (27.3%) and Tebessa-(Sahara), Blida-(AHS) and Bejaia-(AHS) in mineral water production units (10% each).

• •

  The results of this study demonstrate the abundance of mineral elements in most of the Algerian bottled waters.

• •

  All water brands comply with national, European and WHO norms for the bottled mineral and spring waters, except for the (Brand #63 or 3%) in which NO₂⁻ concentration exceeded the maximum permissible limit for mineral water. In addition, TDS and TH values (Brands #4 or 3% for TH and #21 or 3% for TDS) exceeded the recommended guidelines for drinking spring water.

• •

  Nearly 5% of the total brands were of bicarbonate nature belonging to mineral water; however around 25% of all brands were suitable for low sodium diet, recommended for individuals with hypertension.

• •

  From the findings of the study, it is obvious that TDS is the most discriminating characteristic in the categorization of water brands.

• •

  Chadha categorization showed water classes, for the spring water: (I) Ca + Mg-HCO₃ (66.7%), (II) Ca + Mg-Cl + SO₄ (27.3%) and (III) Na + K-Cl types (6%), and for the mineral water: (I) Ca + Mg-HCO₃ (80%), (II) Ca + Mg-Cl + SO₄ (13.3%) and (III) Na + K-HCO₃ types (6.7%). This shows that TH (I + II) alone has reached around 94% for each type of bottled water.

• •

  Using the HCA approach, two main clusters were identified. The first main cluster (G1) contained the majority of the studied brands (62 brands, equivalent of 98%), which could be classified, according to EU Directive 2009/54 EEC, in the range of low to intermediate mineral content waters; and the second cluster (G2) embraced only one brand (#40) with extreme values of TDS, TH, HCO₃^–, Na⁺ and Ca²⁺. So, it is highly rich in mineral salts, very hard and bicarbonate; this makes it suitable for therapeutic purposes.

• •

  In all groups of age and gender taken into consideration, the key elements having significant contributions to DRI were:

• • •

    K⁺ intake is low in both bottled water types,

  • •

    Ca²⁺, Mg²⁺, Cl⁻ and Na are consumed moderately in all bottled spring water brands,

  • •

    Ca²⁺, Mg²⁺, Cl⁻ and Na⁺ are consumed considerably in some of bottled mineral water brands,

  • •

    Na⁺ and Ca⁺² may pose health risks in some of bottled mineral water brands.

Our results suggest the following:

• Not all mineral water is suitable for daily consumption, so consumers should be careful about drinking water with high concentration of microelements.

• Monitoring regularly the quality parameters including minor elements would be helpful to improve the water quality for consumers' safe use.

• Assessing the existing problem of DRI while considering factors including physical activities and different deficiencies.

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

The authors declare there is no conflict.