Finding a relationship between physicochemical characteristics and ionic composition of River Nworie, Imo State, Nigeria

Study area

Nworie River is the study area (Fig. 1). The river originates from Mbiatolu LGA of Imo State between latitude 5°28N and 5°31N (Fig. 1). It passes through Owerri Municipal of Imo State and then empties into the Otamiri river at Nekede, Owerri West LGA, Imo State. The measured length of the river is approximately 9 kilometers. Photographs showing the decline state and anthropogenic activities of some points along Nworie River are shown in Figs. 2A–2D and Table 1 respectively.

Collection of water sample

A total of 30 sub-samples were collected to make five composite (six subsamples per sampling location) samples. Samples were collected following a ‘W’ shaped design along the longitudinal course of the river using the grab technique (Verla et al., in press). From where the river enters the Owerri municipal to where it leaves. Geographically, the sampling points lies between the latitudes 05.52°N and longitude 07.03°E, NDW1 and NDW2 (Amakohia-Alvan), NDW3 (HolyGhost college), NDW4 (Wetheral) lies between 5.479°N and longitude 07.027°E, and NDW5 (river leaving the town) (Fig. 1). The water samples were collected during the dry season period. The points of sample collection were at least 300 m apart, done in morning against the water current. Clean plastic bottles were used for the collection.

Analytical procedure of water sample

The electrical conductivity was measured using HANNA HI8733 EC METER in μS/cm, which was calibrated using KCl. The electrical conductivity (EC) probe was dipped into the water sample for 60 s and readings were recorded on the meter screen.

The pH was determined using JENWAY 3510 pH METER which was calibrated using buffer 4 and buffer 7 by dissolving one capsule each in 100 ml of distilled water respectively. The pH was determined by introducing the probe of the pH meter into water sample (collected from a large sample) and the reading on the meter screen was recorded.

Dissolved oxygen (DO) concentrations were determined with a Jenway 9071 DO analyzer by inserting the probe into the water samples.

TDS (Total Dissolved Solids) and TSS (Total Suspended Solid) (mg/L) was determined according method described by Verla et al. (2018) while phosphate (PO₄³⁻), nitrate (NO₃²⁻), chloride was determined according to American Public Health Association method 4110 (APHA, 2005).

Determination of heavy metals and macro elements

The water sample was digested using aqua-regia (HCl+HNO₃ in ratio 3 to 1). one mL of the water sample was digested for 3 days in a test tube with 24 mL of aqua-regia (Enyoh, Verla & Egejuru, 2018; Verla et al., 2018; Verla et al., 2019a). The digested filtrates were used for the total metal quantification of Na, K, Mg, Ca, Mn Pb, Cd, Fe, Zn, and Cu using Atomic Absorption Spectrophotometer (Perkin Elmer AAnalyst 400) in mg/l (ppm). The characteristic wavelengths of metals determined were first set using the hollow cathode lamp, then digested filtrates samples was aspirated directly into the flame. To ensure accuracy of data, calibration of the equipment was done for each element using a standard sample prepared as a control with every set of samples. The instrumental parameters for particular metals that were analysed are presented in Table 2. The same procedure was done for all five composite samples.

Data analysis

The data were analyzed for basic descriptive statistics such minimum, maximum, mean and standard deviation of triplicate analysis and level of significance was determined using Microsoft Excel and IBM SPSS statistics version 20. Correlation, Principal Component and linear regression analysis were conducted to establish relationship between physicochemical properties and ionic composition at 5% level of significance. P-values were considered significant when less than 0.05. To interpret the r-value for the correlation and linear relationship, the following classification was adopted, presented in Table 3. Hierarchical Cluster Analysis (HCA) was used also to find similarities between physicochemical properties and ionic composition.

Surface water characteristics

The surface water characteristics are presented in Table 4. The results are compared with World Health Organization standards permissible limit (WHO, 2007). The mean values of temperature and total dissolved solids (TDS) exceeded the recommended limit. Mean pH (5.62) was below the acceptable range while other physicochemical parameters didn’t exceed the permissible limit. However, the highest and lowest temperature was recorded at NDW4 and NDW1 respectively. The overlying water in NDW4 had the highest DO value of 3.89 mg/L and in NDW5 the lowest was recorded (2.27 mg/L). The standard DO expected according to WHO permissible limit (2007) is 4 mg/L, hence suggesting relatively poor water quality of Nworie River.

Ionic composition and distribution

The results for the ionic compositions and percentage distributions are presented in Table 5 and Fig. 3. Only calcium, iron, nitrate, phosphate and chloride showed lower mean concentrations when compared to WHO permissible limits (Table 5), all others showed higher mean concentrations. The order of ionic composition was Cl⁻ >SO₄²⁻ >Ca²⁺ >Zn²⁺ >Mg²⁺ >NO₃⁻ >PO₄³⁻ >Na ⁺ >Mn²⁺ >others. Chloride, calcium and zinc showed the highest distributions for anions, cations and heavy metals respectively.

Ion balancing of River

When a water quality sample has been analysed for the major ionic species, one of the most important validation tests can be conducted: the cation-anion balance (HPTM, 1999). The principle of electroneutrality requires that the sum of the positive ions (cations) must equal the sum of the negative ions (anions). Thus the error in a cation-anion balance can be written as (1): (1)${\displaystyle \text{\%}Balanceerror=\frac{\sum Cationsintheriver-\sum Anionsintheriver}{\sum Cationsintheriver+\sum anionsintheriver}.}$

The computed cation-anion balance is presented in Fig. 4. For surface water, the % error should be less than 10. Here, a balance error ranging from 30 to 39.42% is recorded for the different sampling point of the river. The high balance errors suggest that major ionic concentrations distributed in Nworie River are not equal (i.e major cations ≠ major anions).

Correlation analysis and Principal component analysis of various ions

The result for the Pearson’s correlation analysis of various ions in Nworie river is presented in Table 6. This model has been used well by many researchers in determining contamination sources for pollutants in the environment (Enyoh, Verla & Egejuru, 2018; Duru, Okoro & Enyoh, 2017; Verla, Verla & Enyoh, 2017; Ibe, Isiukwu & Enyoh, 2017). The value of r is always between +1 and –1. Most ions exhibit strong relationships, suggesting that their presence in the water is from similar anthropogenic source(s). Strong relationship was exhibited by some of the ions. Cations exhibited strong relationship amongst them. Na further showed strong relationship with some heavy metals (Cu, Cd, Mn and Fe) and anion (PO₄³⁻). Similar relationships were also exhibited Mg²⁺ and Ca²⁺. Strong relationship was exhibited between heavy metals amongst them and with Cl⁻ and PO₄³⁻, except for Zn. Most anions showed negative relationships except for NO₃⁻ / SO₄²⁻(0.54). To determine the precise contamination source(s) of ions as predicted by the correlation analysis; we conducted a principal component analysis following standard procedures (Dragović, Mihailović & Gajić, 2008; Franco-Uría et al., 2009). We used the varimax rotation with Kaiser Normalization because it better explained the possible groups or sources that influence the soil system and maximise the sum of the variance of the factor coefficients (Gotelli & Ellinson, 2013). The factor loadings for heavy metals in the river water are shown in Table 7. Three components were extracted based on eigen value >1, best described the sources.

Relationship between water characteristics and dissolved ions

In order to establish a relationship between the recorded physicochemical properties and ionic composition of Nworie river, multiple linear regression analysis was conducted. Linear regression is a statistical method used to assess a possible linear association between two continuous variables. The analysis gives out a regression coefficient (R² value) (Enyoh, Verla & Egejuru, 2018). R² value reveals the extent of influence of river physicochemical properties on the distribution of ions in the river on the basis of closeness to either –1 or +1 to indicate a strong enough linear relationship. The computed relationship for cations, anions and heavy metals are presented in Figs. 5–7. All ions showed positive relationships at varying level with the water physicochemical characteristics.

Hierarchical Cluster analysis

The result for Hierarchical cluster analysis (HCA) is presented in Fig. 8, most the ions showed similarities with measured physicochemical parameters. Only TDS and EC showed some dissimilarity with iron and sulphate.

Surface water characteristics

Only mean values of temperature and total dissolved solids (TDS) exceeded the recommended limit (Table 2). The characteristics of the river are a reflection of the sampling period. High temperature obtained wasn’t surprising due sampling season (dry season). Low values of DO have been reported earlier in other studies (Manila & Frank, 2009; Duru & Nwanekwu, 2012; Okoro, Uzoukwu & Ademe, 2016; Verla et al., in press; Verla et al., 2019a). These studies related the low DO to human activity, causing enrichment of the surface water with high organic content. Electrical conductivity recorded in this study fell below the acceptable limit (100 µS/cm) set by WHO (2007). The conductivity depends on water temperature and is the measure of water capability to pass electric flow. High temperature may cause high EC. The conductive ions may come from dissolved salts and inorganic materials in the River.

Ionic composition and distribution

Dissolved salt often exists as ions in solution. The ions can be positive or negative. Ions with a positive charge are called “cations” while negative charge is called an “anion”. Going by this definition, heavy metals are cations because they are positively charged. However in this study we classified the ions as major cations, heavy metals (often categorized as secondary constituents) and anions. The major cations studied were sodium, potassium, magnesium and calcium; heavy metals were copper, cadmium, manganese, zinc, iron and lead; while anions were nitrate, sulphate, phosphate and chloride. The results for the ionic compositions are presented in Table 4. The major cationic and anionic concentrations were low generally when compared to WHO permissible limit (2007) standard except for magnesium. The magnesium concentration ranges from 1.13 mg/L to 2.78 mg/L as opposed to the WHO permissible limit (2007) standard of 0.5 mg/L. Calcium (Ca²⁺) and Magnesium (Mg²⁺) ions are both common in natural waters and both are essential elements for all organisms (HPTM, 1999). They are responsible for the hardness of natural waters when combined with dissolved materials in the water. WHO reported that hard water has no known adverse health effect (Sengupta, 2013) and could provide an important supplementary contribution to total calcium and magnesium intake (Galan et al., 2002). However, prolong consumption of water with high magnesium can cause hypermagnesemia if the consumer have significantly decreased ability to excrete magnesium (Chandra et al., 2013).

All heavy metals except for Fe had mean values exceeding the permissible limit set by WHO (Table 5), which is indicative that the water is polluted by these metals. The concentration of Pb ranged from 0.127 to 0.521 mg/l, Fe ranged from 0.091 to 0.191, Cd ranged from 0.002 to 0.180 mg/l, Cu ranged from 0.13 to 0.79 mg/l, manganese ranged from 0.08 to 1.02 and zinc ranged from 1.2 to 2.63 with mean values of 0.22, 0.13, 0.05, 0.35, 0.41 and 2.25 mg/l respectively. The solubility of trace metals in surface waters is predominantly controlled by the water pH, the type and concentration of ligands on which the metal could adsorb, and the oxidation state of the mineral components and the redox environment of the system. Ingestion of metals such as lead (Pb) and cadmium (Cd) may pose great risks to human health by interfering with essential nutrients in the body, possibly causing small increases in blood pressure and damaging the kidney. In addition, they can equally affect aquatic fauna and flora (Isiuku & Enyoh, 2019).

The percentages of ions are presented in Fig. 3. This distribution is showing the abundance of major ions in Nworie river during dry season. Metal ions and anions highly distributed were calcium (26.07%), sulphate (26.90%), chlorine (34.88%), while sodium, potassium, nitrate, phosphate, zinc were within the range of 1–3%. Other heavy metal ions showed distribution of 0%. These suggest that the presence of heavy metals in Nworie river during dry season is low in the total composition of ions in the River. The low distribution could probably due to the metals forming complexes with organic materials in the water or they could be in abundance in other forms which can be accessed through chemical speciation (Verla et al., 2019c).

Ion balancing of River

The ionic balancing of the river is high (mean balance error of 35.62%) indicating variable dissolution of ions in the river (Fig. 4). The obvious reason could be due to pollution by anthropogenic means through dumping of waste and the river finding itself at the receiving end of an effluent discharge from domestic and industrial sources. Fertilizers from farmlands along Nworie river finds their way into the river through surface run-off and increases the anionic concentrations such as PO₄³⁻ and NO₃⁻ in the river while major cations might undergo complexometric reactions reducing their concentrations in the water. Anthropogenic activities have been reported to significantly alter ionic concentration in water (Manila & Frank, 2009; Enyoh, Verla & Egejuru, 2018; Verla et al., 2018; Isiuku & Enyoh, 2019).

Correlation analysis and Principal component analysis of various ions

The result for the Pearson’s correlation analysis of various ions in Nworie river showed many ions showing strong relationships suggesting that their presence in the water is from similar anthropogenic source(s). Some association exhibited by the ions has observed in other water studies (Enyoh, Verla & Egejuru, 2018; Duru, Okoro & Enyoh, 2017; Verla, Verla & Enyoh, 2017; Ibe, Isiukwu & Enyoh, 2017). Precise contamination source of ions was determined by PCA. Three components were extracted based on the eigen value greater than 1. The three components cumulatively explained 93.324% of the total variance and generally indicated an anthropogenic source(s) of the studied ions in the river water (Table 7). According to Liu, Lin & Kuo (2003), components loadings values of >0.75, 0.75–0.50, and 0.50–0.30 were classified as “strong”, “moderate”, and “weak respectively. The PC1 explained 62.424% of total variance and was found to be strongly and positively correlated to Cd²⁺, Cu²⁺, Ca²⁺ , Mn ²⁺ , Na⁺ , Fe²⁺ , PO₄²⁻ and Mg²⁺ (0.71–0.99). This relates to the artisanal activities, metal processing works in the area and agricultural activities; PC2 explained 16.088% of total variance and showed moderate to strong positive factor loadings for SO₄ ²⁻ and Pb (0.65–0.90) which also indicate industrial source, activities of scrap metal dealers, recycling of metals as well as lead–acid accumulators, vehicle emissions in urban areas and agricultural activities in the study area. PC3 explained 14.812% of total variance and showed strong positive factor loadings for NO₃⁻ and K⁺ (0.72–0.81) which could be attributed to atmospheric deposition and agricultural activities involving the use of fertilizers.

Relationship between water characteristics and dissolved ions

The computed relationship for cations, anions and heavy metals are presented in Figs. 5–7.