Introduction

River water aids human development, and various forms of life depend on it for support and maintenance (Bhardwaj et al. 2017). It generally serves as a source of potable water where groundwater supplies are inadequate/unavailable (Seiyaboh et al. 2017), and it naturally possesses a high degree of self-purification. However, the latter is disturbed by excessive pollution problems, especially those occasioned by socio-economic, institutional, and industrial processes (Mbuligwe and Kaseva 2005). Such “anthropogenic” activities, as well as natural processes, can introduce heavy metals, polycyclic aromatic hydrocarbons, pesticides, and polychlorinated biphenyls, among others, which can impair water quality (Ali et al. 2016; Otene and Alfred-Ockiya 2019). Heavy metals, when present in river water above the safe limits, are well-known to be a major threat to man, mainly because such rivers provide water for human consumption, and a habitat for aquatic organisms (Ochieng et al. 2008).

Contamination of aquatic ecosystems by metals is a foremost issue due to the abundance, non-biodegradability, persistence, toxicity, and bioaccumulation of metals (Shinjo et al. 2014). The bioaccumulation of metals in the organs and tissues of aquatic species can lead to various health consequences that transcend to man, primarily when these species are utilized as a long-term source of nutrition. In Nigeria’s Niger Delta area, crude oil spills account for a large proportion of the heavy metal contamination of rivers (Vincent-Akpu et al. 2015). Although these contaminants are not easily noticed relative to several other kinds of water pollutants due to their great solubility, their low rate of metabolism suggests that removal through self-purification processes will be negligible (Harikumar and Nasir 2010; Osioma and Iniaghe 2019).

In Nigeria, the presence of heavy metals in rivers at elevated concentrations is attributed to leachates from dumpsites, industrial discharges, and corrosion of iron, among other factors (Ayenimo et al. 2005). Pollution of river water by heavy metals was first reported by Monbreshola et al. (1983). Over time, several studies have been conducted on various rivers across Nigeria with a focus on determining the overall water quality in respect of the contaminant concentrations. Oboh and Agbala (2017) and Egun and Ogiesoba-Eguakun (2018) reported the excellent quality of water in the Siluko and Okhuaile rivers of Edo State Nigeria. Severe contamination of the Ikpoba reservoir was reported by Egun and Oboh (2021), and heavy metals, such as Pb, Cd, and Fe, were identified as major contributors to the deteriorated water quality. Atalawei et al. (2020) reported that water from the Olugbobiri and Ogboinbiri creeks in the Niger Delta was not suitable for drinking, owing to the high levels of Fe and polycyclic aromatic hydrocarbons in them. In a related report, Nduka and Orisakwe (2011) showed that the concentrations of several heavy metals determined in selected rivers and creeks in the Delta and Rivers States in the Niger Delta were above permissible limits.

Water quality often refers to the physical, chemical, biological, and radiological characteristics of water (Kondum et al. 2021). The water quality index (WQI),  and comprehensive pollution index, are reliable indices for providing useful information on water quality. The WQI is a unit-less number, and it expresses the overall quality of water as a single number by classifying it according to its suitability for domestic purposes (Tyagi et al. 2013). Multivariate statistical analysis can establish likely sources of contaminants in such water bodies.

The Bomadi Creek is a major offshoot of the Forcados River. It serves the purpose of providing drinking water as well as transportation, fishing, sand dredging, recreation, and even waste disposal. The river water flows through various settlements, with different degrees of anthropogenic activities, which can influence the overall water quality. Although several studies have been conducted on the quality of surface waters in the Niger Delta, there is a lack of data on the water quality in Bomadi Creek. Therefore, the objectives of this study were to determine some physicochemical properties and metal concentrations of surface water from Bomadi Creek with a view to providing data on the quality status, sources of contamination, and risk to humans. Information from such a study is helpful in designing water treatment schemes and for water quality management.

Materials and methods

Description of study area

The study area was chosen to evaluate the impact on the overall water quality of oil exploration and exploitation activities within and around the river, as well as the activities of different human settlements situated along the river banks. The area fell within an approximately 45 km stretch of the river, from upstream (Bomadi) to downstream (Ndoro) sections of the river. Sampling locations labeled S1–S3 were located downstream; S4–S6 were midstream, while S7 and S8 were upstream. A map of the study area is shown in Fig. 1 and a description of the points sampled follows.

Fig. 1
figure 1

Map of the study area showing the sampling locations

Full size image

S1 was located at Ndoro town, close to a bread-making factory, whose wastes are directly dumped into the river.

S2 was a sparsely populated settlement, located about 5 km upstream of S1, with heavy presence of water hyacinth.

S3 was located close to Gbese Grammar School, Ojobo. It was about 500 m from the Benisede flow station located at an adjoining creek (Peretorugbene Creek). There was also a heavy presence of water hyacinth.

S4 and S5 were densely populated human settlements, located at Ojobo and Torugbene communities, respectively. Conspicuous activities in these locations included laundry and small-scale industrial activities.

S6 was located close to a mobile police stop and check post, with remarkable human activities, at Ekeremor town. A petrol station was also situated about 50 m upstream from this station.

S7 and S8 were located at Tuomo town and Bomadi overside, respectively. The latter was about 50 m from the bridge constructed across the Forcados River. These stations were characterized by dense human settlements.

Sampling

Water samples were collected from the aforementioned locations with the aid of 1-L pre-cleaned plastic containers at a depth of 2–10 cm below the surface of the water, between 7h00 and 9h00, and across two seasons—the dry season (February–April) and the wet season (August–October). Sample containers were initially soaked in 5% nitric acid overnight, rinsed with deionized water, and then with the river water. The water samples were acidified with 2 mL of conc. nitric acid, capped, and stored at 4 °C (Singh et al. 2005).

Analysis

The pH values of the samples were determined in-situ with the help of a hand-held digital pH meter (pHep®, Hanna, USA). Total suspended solids (TSS) and total dissolved solids (TDS) were estimated gravimetrically. Dissolved oxygen (DO), biochemical oxygen demand (BOD5), nitrate (NO3ˉ), hardness, and turbidity were determined by standard methods (APHA 2017).

For the metals analysis, samples were digested with nitric acid and quantified by means of atomic absorption spectrophotometry (PerkinElmer Model 3110). To 100 mL of a water sample contained in a 250-mL conical flask, 10 mL of 1 M nitric acid (90%, Sigma-Aldrich) was added. The flask was placed in a digestion block and heated at 150 °C until the mixture was reduced to about 25 mL. Then, it was removed from the block, cooled, filtered through a Whatman No. 40 filter paper into a 100-mL volumetric flask and made up to the mark with 1 M nitric acid (Csuros and Csuros 2002). The equipment was calibrated with analytical standards of the respective metals. These standards (1000 mg L−1) were serially diluted to obtain the working concentrations and, subsequently, after measurement, a calibration graph was constructed.

Quality control and statistical analysis

The acids used for the analyses were of analytical grade. Procedural blanks were used to correct the results for any impurities in the reagents/vessels. All samples were analyzed in triplicate and the relative standard deviations (RSD) for the triplicate analyses were < 10%. For metal analysis, the method was validated by using a spike recovery. The percent recoveries for the metals ranged between 89.2 and 101%. Data were expressed as the mean ± standard deviation. One-way analysis of variance (ANOVA) was used to determine whether the properties determined varied significantly across the various sampled locations, with a p-value less than 0.05 (p < 0.05) taken to be significantly different. Principal component analysis (PCA), Pearson’s correlation analysis, and hierarchical cluster analysis were performed on the generated data to provide useful insights on the possible sources/causes of pollution in the studied creek by reflecting on the interrelationships that existed among the pollutants under study. PCA is based on a varimax rotation technique which is used to investigate whether a number of variables of interest are linearly related to a smaller number of unobservable factors (Fatai 2011). The statistical calculations were performed with SPSS version 11.5.

Data evaluation

Water quality index

The water quality index (WQI) was evaluated by using the weighted arithmetic water quality index method (Egun and Oboh 2021; Egun and Ogiesoba-Eguakun 2018; Oboh and Agbala 2017; Tyagi et al. 2013). The reference standard used for the computation was the Nigerian standard for drinking water quality (SON, 2007). The WQI was computed from Eq. (1):

$$\mathrm{WQI}=\frac{{\sum }_{i=1}^{n}{W}_{i}{Q}_{i}}{{\sum }_{i=1}^{n}{W}_{i}}$$
(1)

where Wi is the unit weight of the ith parameter, Qi is the quality rating scale of the ith parameter, and n is the number of parameters considered.

The unit weight Wi is given by Eq. (2):

$${W}_{i}=\frac{K}{{S}_{i}}$$
(2)

where Si is the maximum allowable recommended standard for the ith parameter, and K is the proportionality constant, given by Eq. (3):

$$K=\frac{1}{\sum (1/{S}_{i})}$$
(3)

The quality rating, Qi, was obtained by using the expression given in Eq. (4):

$${Q}_{i }=\sum_{i=1}^{n}\frac{{M}_{i}{-I}_{i}}{{S}_{i}-I_{i}}\times 100$$
(4)

where Mi is the experimental value obtained for the ith parameter, Ii is the desirable/ideal value of the ith parameter in pure water (Ii = 0 for all parameters except pH (with a value of 7.0) and dissolved oxygen (with a value of 14.6 mg L−1)), and Si is the recommended standard value of the ith parameter. The quantity \({M}_{i}{-I}_{i}\) is the numerical difference of the two values without regard to the algebraic sign.

WQI values are categorized into five classes based on the weighted arithmetic water quality index method (Tyagi et al. 2013) as follows: 0–25 (excellent water quality), 26–50 (good water quality), 51–75 (poor water quality), 76–100 (very poor water quality), and > 100 (water unsuitable for drinking).

Comprehensive pollution index

The comprehensive pollution index (CPI) of the water samples was calculated by using the single factor pollution index (PI) as described by Eq. (5):

$${P}_{i}=\frac{{C}_{i}}{{S}_{i}}$$
(5)

where \({P}_{i}\) is the pollution index of pollutant i; \({C}_{i}\) is the measured concentration of the pollutant (mg L−1); and \({S}_{i}\) is the national water quality standard permissible limit for the pollutant in surface water (SON, 2007). The water quality factor \({P}_{i}\) is classified into five grades as shown in Table 1 (Yan et al. 2015).

Table 1 Standard grades for single factor pollution index (PI) (Yan et al. 2015)
Full size table

The CPI is the arithmetic mean of n water pollution indices as given by Eq. (6):

$$CPI=\frac{1}{n}\sum \frac{{C}_{i}}{{S}_{i}}$$
(6)

where \({C}_{i}\) is the measured concentration of the pollutant (mg L−1), \({S}_{i}\) is the national water quality standard permissible limit for the pollutant in surface water (SON, 2007), and n is the number of analyzed pollutants. The CPI is classified according to the water quality levels listed in Table 2.

Table 2 Standard grades for comprehensive pollution index (CPI) (Yan et al. 2015)
Full size table

Human health risk assessment of heavy metals in water

The non-cancer and cancer risk assessment from pollutants in water was evaluated as the hazard index and total cancer risk, respectively. The hazard index and total cancer risk were evaluated for the two major exposure routes: ingestion and dermal contact with water. The evaluation was carried out by using Eqs. (7), (8), (9), (10), (11), and (12) (USEPA 1989; USEPA, 2009).

For the non-cancer risk,

$$\mathrm{Hazard\, index }\hspace{0.17em}(\mathrm{HI})\hspace{0.17em}=\hspace{0.17em}\sum \mathrm{HQ}={\mathrm{HQ}}_{\mathrm{ing }}+{\mathrm{HQ}}_{\mathrm{dermal}}$$
(7)
$$\mathrm{HQ}=\frac{{\mathrm{CDI}}_{\mathrm{nc}}}{\mathrm{RfD}}$$
(8)
$${\mathrm{CDI}}_{\mathrm{ing}-\mathrm{nc}}=\frac{\mathrm{C} \times \mathrm{ IngR}\times \mathrm{ EF}\times \mathrm{ED }}{\mathrm{BW} \times{\mathrm{AT}}_{nc}}\times {10}^{-6}$$
(9)
$${\mathrm{CDI}}_{\mathrm{dermal}-\mathrm{nc}}=\frac{\mathrm{C }\times \mathrm{ SA }\times \mathrm{ Kp }\times \mathrm{ ET }\times \mathrm{ EF }\times \mathrm{ ED }\times \mathrm{ CF }}{\mathrm{BW }\times {\mathrm{AT}}_{nc}}$$
(10)

For the cancer risk,

$${\mathrm{Risk}}_{\mathrm{ing}}=\frac{\mathrm{C} \times\mathrm{IngR} \times \mathrm{EF} \times \mathrm{ED} \times \mathrm{CF} \times \mathrm{SFO}}{BW \times {AT}_{ca}}$$
(11)
$${\mathrm{Risk}}_{\mathrm{dermal}}=\frac{\mathrm{C} \times \mathrm{SA} \times \mathrm{Kp} \times \mathrm {ET} \times \mathrm{EF} \times \mathrm{ED} \times \mathrm{CF} \times \mathrm{SFO} \times \mathrm{GIABS}}{\mathrm{BW} \times\mathrm {AT}_{ca}}$$
(12)

where CDIing is the chronic daily intake for ingestion, CDIdermal is the chronic daily intake for dermal contact, Risking is the risk for ingestion, and Riskdermal is the risk for dermal contact.

The evaluation of the hazard index and total cancer risk was carried out by using the parameters listed in Tables 3 and 4, respectively. A hazard index value less than 1 indicates there is no adverse non-carcinogenic risk, while a hazard index value greater than 1 indicates the presence of adverse non-carcinogenic risks. A total cancer risk value less than 1 × 10–6 indicates no cancer risk but when greater than 1 × 10–6 it indicates a cancer risk (USEPA, 2010).

Table 3 Definitions and values of variables for human health risk assessment
Full size table
Table 4 Toxicological parameters for the investigated pollutants used for human health risk assessment
Full size table

Results and discussion

Physicochemical characteristics

A summary of the values of some physicochemical properties of the water from Bomadi Creek is presented in Table 5. The mean pH values, which ranged from 7.40 to 8.17, indicated slight alkalinity of the water, but the values were within the regulatory limit of 6–9 stipulated by the World Health Organization (WHO, 2006). In general, the wet season samples recorded higher pH values than the dry season, but such variations were not significant (p > 0.05). This could be attributed to the buffering capacity of the river. In addition, there was a steady increase in the pH values from upstream to downstream locations, which could be due to the incursion of highly saline water from the Atlantic Ocean, which was closer to the downstream sampling points than the upstream ones. Similar alkaline pH values have been previously recorded for waters from Bodo Creek (Vincent-Akpu et al. 2015), parts of the Warri River (Tesi et al. 2019), and the River Benue (Iwar et al. 2021a), while near-neutral to acidic pH values were reported in creeks from Sagbama (Seiyaboh et al. 2017) and Kolo (Aghoghovwia and Ohimain 2014; Eremasi et al. 2015), respectively.

Table 5 Physicochemical properties of river water from Bomadi Creek
Full size table

The mean concentration of TDS ranged from 27.5 to 44.7 mg L−1. The dry season recorded higher TDS levels than the wet season, and the difference was significant (p < 0.05). The concentration of TDS in river water depends on the geological materials that the water passes through, as well as the quality of the infiltrating water (Oram 2014). The low TDS values recorded in the wet season could be attributed to the impact of extrinsic factors such as floodwater associated with increased rainfall, which consequently resulted in dilution effects. The TDS values were observed to be lower than the WHO limit of 1000 mg L−1. Similarly, the water can be also be classified as fresh water, since its total TDS was less than 5000 mg L−1 (Ademoroti 1996).

For TSS, the mean concentration ranged from 34.1 to 61.7 mg L−1 across both seasons. Unlike TDS, the dry season recorded significantly lower concentrations (p < 0.05) than the wet season. The high influxes of materials through run-off from the riparian zone, and agitation of particulate organic matter already settled at the bottom of the river, are likely factors responsible for this observation (Ikomi and Emuh 2000). High concentrations of suspended solids can cause several problems for rivers, streams and aquatic lives (Iwar et al. 2021a, b).

For turbidity, the mean values, ranging from 20.5 to 42.3 NTU, exceeded the Standards Organization of Nigeria (SON, 2007) guideline value of 5 NTU in water used for domestic purposes. The difference in turbidity values for both seasons was significant, with the wet season having higher levels in all locations. Increased turbidity in the wet season was assumed to be due to the influx of increased silt and particulate substances into the water due to surface run-off from settlements. This trend has been previously reported (Ikomi and Emuh, 2000; Fatoki et al. 2002).

The DO levels ranged from 3.07 to 6.53 mg L−1. The highest levels were found in locations S2 and S7, during the wet season, while the lowest level was found in S6 for both seasons. The low levels recorded in S6 could be linked to activities from the petrol station located close by—petroleum products were seen floating on the water surface, and this could reduce aeration of the water and, subsequently, DO. The mean DO levels were higher in the wet season than the dry season, but the difference was not significant (p > 0.05). Similar seasonal variations have been reported earlier (Imoobe and Oboh 2003). The DO levels in all locations except S6 indicated that the river water was well aerated. This was expected, since the water was not static, and there was direct diffusion at the surface, and various forms of water agitation, such as wave action and turbulence. Similar DO values have been reported in water from Bodo (Vincent-Akpu et al. 2015) and Sagbama (Seiyaboh et al. 2017) Creeks. The acceptable minimum DO level that can maintain a fish population in the aquatic environment is between 4 and 5 mg L−1, while fish mortality reportedly occurs when DO levels are less than 3 mg L−1 (Ovie and Adeniji 1990; Oram 2014). Thus, the DO levels in Bomadi Creek are at acceptable levels.

The biochemical oxygen demand (BOD) levels in the river water were generally low, ranging from 0.13 to 2.77 mg L−1. The BOD levels were relatively higher in the wet than in the dry season, but the difference was not significant (p > 0.05). With respect to the BOD levels and the aquatic pollution status of waters, BOD concentrations < 1.0 mg L−1 are classified as unpolluted, 2 ≤ BOD ≤ 9 mg L−1 are classified as moderate pollution, and BOD > 10 mg L−1 are classified as heavy pollution (Vowels and Connel 1980; Mara 1983; and Adakole et al. 1998). Similarly, the maximum permissible limits set by the Department of Petroleum Resources (DPR 2002) and the WHO (WHO 2005) are 10 mg L−1 and 5.0 mg L−1, respectively. The water could therefore be said to be moderately polluted with organic matter.

The mean nitrate levels ranged from 0.37 to 0.92 mg L−1 for both seasons and were all within the Federal Enviromental Protection Agency of Nigeria safe limit of 10 mg L−1 for nitrates in water. The low levels could be due to the near-absence of agricultural activities in the studied locations.

For total hardness, the mean concentration ranged from 28.3 to 60.0 mg L−1 CaCO3. This relates to soft and moderately soft water (Fatoki et al. 2002). Higher values were observed in the dry season than the wet season, but the difference was not significant (p > 0.05). Although no health-based guideline exists for water hardness, concentrations exceeding 200 mg L−1 may be responsible for scale deposition in distribution systems and will also result in excessive soap consumption and, subsequently, scum formation (WHO 2003). Also, lead and cadmium toxicities are reported to vary with water hardness (Fatoki et al. 2002).

Seasonal variation in heavy metal concentrations

A summary of the seasonal variation in the heavy metal concentrations in the water samples is shown in Table 6. The average metal concentrations for both seasons followed the order: Fe > Zn > Cu > Mn > Cd > Pb. The mean concentration of Pb in the water samples ranged from 0.01 to 0.02 mg L−1 for both seasons. There was no clear pattern in seasonal variation, neither was there any observed significant variation (p > 0.05) in concentration for the two seasons. The observed Pb concentrations were greater than the permissible limit of 0.01 mg L−1 set by the SON (SON 2007) and the WHO (WHO 2011). This suggests a potential health risk associated with this water when used for drinking purposes. The results were similar to reports from Kolo Creek (Aghoghovwia and Ayatari 2012; Eremasi et al. 2015), but lower than values reported for Bodo Creek (Vincent-Akpu et al. 2015), and parts of Warri (Tesi et al. 2019) and Ethiope Rivers (Agbaire and Obi 2009), and higher than reports for Elechi and Ikoli Creeks (Otene and Alfred-Ockiya 2019; Ighariemu et al. 2019).

Table 6 Metal concentrations in water from Bomadi Creek
Full size table

The mean concentration of Cd in the water samples ranged from 0.01 to 0.03 mg L−1. The highest levels were observed in S1 and S7. The concentration of Cd obtained in this study exceeded the regulatory limit of 0.003 mg L−1, but was within the maximum contaminant level goal (MCLG) of 0.05 mg L−1 for drinking water set by the WHO (WHO, 2006). High Cd levels exceeding the WHO regulatory limit have been reported in parts of Warri River (Tesi et al. 2019), Elechi Creek (Otene and Alfred-Ockiya 2019) and Kolo Creek (Eremasi et al. 2015), while Ighariemu et al. (2019) and Aghoghovwia and Ayatari (2012) reported Cd levels within the permissible limit.

The mean concentrations of Cu in the water samples ranged from 0.03 to 0.07 mg L−1 across both seasons and were all within the WHO guideline value of 2 mg L−1. The values obtained in this study were greater than those reported for Elechi (Otene and Alfred-Ockiya 2019) and Kolo (Eremasi et al. 2015) Creeks. The generally low levels could be attributed to the fact that Cu is not a raw water contaminant.

The mean Fe concentrations varied from 0.58 to 0.94 mg L−1. Apart from location S1, the wet season had higher Fe values than the dry season. The highest concentration was recorded in location S8, during the wet season, and the elevated level could be attributed to scrap metal indiscriminately dumped in and around the river at this stretch. Although there is no guideline value for Fe in drinking water proposed by the WHO (WHO 2006), the SON (SON 2007) specifies a limit of 0.1 mg L−1 and a maximum allowable level of 1 mg L−1. High Fe levels in water can result in the development of taste, color and other aesthetic problems (Fatoki et al. 2002). Similar results were reported for Sagbama Creek (Seiyaboh et al. 2017), Siluko River (Oboh and Agbala 2017), and Ethiope River (Ogbeibu and Anagboso 2003; Omo-Irabor and Olobaniyi 2007).

For Zn, the concentrations ranged from 0.11 to 0.25 mg L−1. Although Zn levels in surface waters do not generally exceed 0.01 mg L−1 (WHO 2006), the increased levels of Zn in water in this study can be attributed to Zn being a constituent of roofing sheets, which is subsequently leached by rainfall, and ends up in the river via surface run-offs. However, these concentrations were still within the permissible limit of 3 mg L−1 set by the WHO (WHO 2011). The concentrations recorded in this study were higher than the values reported for Ikoli Creek (Ighariemu et al. 2019). Zinc levels exceeding 3 mg L−1 are reported to give an undesirable taste to water (Oyem et al. 2015). The mean concentrations of Pb, Zn, Cd, and Cu in surface water from the Bomadi creek were relatively lower than those of River Benue (Iwar et al. 2021b).

The concentration of Mn in the water samples ranged from 0.01 to 0.05 mg L−1. The reported values are within the concentration of 0.1 mg L−1 acceptable to consumers and the permissible limit of 0.4 mg L−1 (WHO 2006).

Water quality assessment

Water quality index

A water source is considered suitable for drinking and other domestic purposes when its WQI value is less than 50. The WQI values obtained in this study are shown in Table 7. The values ranged from 949 to 1850 in the dry season, and from 972 to 2763 in the wet season, across the studied locations. The elevated WQI values indicate that the water quality at the various study locations is of very poor quality (i.e., grade E), and, therefore, is unsuitable for human consumption and domestic use.

Table 7 WQI values at the various sampling locations of Bomadi Creek
Full size table

Comprehensive pollution index

Table 8 lists the values for the PIs, as well as for the CPIs, for all the determined water quality parameters across the studied locations. A high PI value is indicative of a significant contribution of that parameter/pollutant in the water body (Egun and Oboh 2021). The contributions of Cu, Zn, Mn, nitrate, TDS, and hardness to the pollution status of Bomadi Creek were minimal (PI values < 0.4); the contribution of pH and Pb was medium (PI values between 1 and 2), while that of Fe was heavy (PI values between 2.1 and 5.0). On the other hand, turbidity, BOD, and Cd contributed seriously to the pollution of Bomadi Creek (PI values > 5.0) across both seasons. Small concentrations of toxic metals present in water have been reported to render the overall quality of surface waters to be poor (Bhardwaj et al. 2017). Elevated levels of Pb and Fe are an indication of the potential human health risk associated with consumption of such waters.

Table 8 Pollution index values for various parameters measured at the different sampling locations along Bomadi Creek
Full size table

The CPI values ranged from 2.84 to 5.50 in the dry season, and from 4.03 to 8.29 in the wet season. Based on the classification of CPI (Table 2), the water in Bomadi Creek is categorized in the “serious pollution” grade, with significant contributions from turbidity, BOD, Cd, and Fe; a medium contribution from pH and Pb, and minimal contribution from TDS, hardness, nitrate, Cu, Zn, and Mn, in both the wet and dry seasons. The sources of high loads of Cd, Fe, and turbidity in the river system include discharges from oil production activities, speed boats, illegal bunkering, and dumping of domestic waste. Like the WQI, the CPI values recorded in this study, in both the wet and dry seasons, indicate serious pollution of Bomadi Creek, and consequently, the water is unsuitable for drinking and other domestic purposes.

Multivariate statistical analysis

Correlation analysis

The results of the correlation analysis are shown in Table 9. It was found that DO was moderately and positively correlated with turbidity in both seasons, which suggests that these parameters have similar sources. BOD also moderately correlated with turbidity in the wet season. For metals, no significant correlations were observed, suggesting that most of the metals were introduced into the water from different sources. However, in the dry season, Mn correlated strongly with Fe, and moderately with Zn and Cu, while in the wet season Mn exhibited a weak correlation with Pb, indicating a similar origin of these metals.

Table 9 Pearson’s correlation coefficients for physicochemical properties and metal concentrations of Bomadi Creek
Full size table

Principal component analysis

The PCA for the physicochemical properties and metal concentrations in surface water from Bomadi Creek for the dry season was resolved into five components that captured 92.3% of the variance (Table 10). Factor 1 represented 23% of the variance and was dominated by TDS, hardness, and the Cd and Cu concentrations. This suggested the TDS of the surface water was influenced by contributions from Cd and Cu compounds within the basin. Factor 2 accounted for 21.6% of the variance and was characterized by positive loading values for Cu, Fe, and Mn. This suggests that Cu, Fe, and Mn arise from the same sources. Factor 3 captured 19.4% of the variance and was characterized by high positive loading values for pH and TSS, and negative bipolar loading values for BOD and Pb. This implies the TSS of the surface water was influenced by the pH. Factor 4 explained 16.1% of the variance with positive loading values for turbidity, DO, and BOD. This depicts the relationship between DO and BOD. Factor 5 has a positive loading value for nitrate, and a negative loading for Zn, which suggests that the source of nitrate in the Bomadi Creek is different from those of the other parameters.

Table 10 PCA of physicochemical properties and metal concentrations of surface water from Bomadi Creek
Full size table

Similarly, the PCA for the physicochemical characteristics and metal concentrations in surface water of Bomadi Creek during the wet season was resolved into five components that captured 94.2% of the variance in the data (Table 10). Factor 1 constituted 24.1% of the variance with positive loading values for pH, TDS and Mn, and negative bipolar loadings for turbidity and Fe. Factor 2 was characterized by positive loading values for DO and BOD and explained 19.7% of the variance. Factor 3 accounted for 19.3% of the variance and was dominated by hardness and Zn, and a high negative loading value for Cu. Factor 4 explained 17.6% of the values with a high positive loading value for nitrate, and a negative loading value for Cd. Factor 5 constituted 13.4% of the variance and was characterized by high positive loading values for TSS and Zn.

Cluster analysis

The variations of the analytes at the investigated sampling points were analyzed by hierarchical cluster analysis. The formation of clusters was based on the average linkage between the group and rescaled distance combined. To illustrate the distribution of metals across the sampling sites (Fig. 2A) and their pattern of occurrences (Fig. 2B), clusters were created based on the average metal concentration in the wet and dry seasons as shown in the combined dendrogram in Fig. 2. The sampling sites were initially clustered into two groups, before splitting into five groups for both seasons. The first sub-group contained sites S7DS, S4DS, S6DS, and S2DS. These are samples from the dry season with relatively high concentrations of Cu, Mn, Fe, and Zn. The second sub-group consist of sites with the highest concentration of Fe (S2WS, S8WS and S7WS), while those in the third sub-group (S5DS and S3DS) recorded the second-highest concentrations of Fe. The fourth and fifth groups consist of sites with intermediate total metal concentrations. The dendrogram showing the pattern of metal occurrences (Fig. 2B) was divided into three clusters based on the increasing order of concentrations across all the sampling sites (Fe > Zn > Cu > Mn > Cd > Pb).

Fig. 2
figure 2

Dendrogram showing the distribution of metals across the sampling sites (A) and their pattern of occurrences (B), created from the average metal concentrations in the wet and dry seasons

Full size image

Health risk assessment

The hazard quotient and hazard index values for estimating the non-carcinogenic risk of consuming water from the Bomadi Creek are shown in Table 11. It can be seen that the HQdermal values were less than 1 for both children and adults across the two seasons investigated. The HQing values ranged from 2.66 to 7.28, with an average value of 4.53 for children, and from 0.81 to 2.23, with an average value of 1.38 for adults. The data showed that exposure to metals through ingestion of water is greater than through dermal contact in the studied locations. The calculated HI values varied from 2.84 to 7.81 (mean value of 4.85) for children, and from 0.88 to 2.43 (mean value of 1.51) for adults. This shows that children are more susceptible to adverse non-carcinogenic risks in the studied locations than adults, and this is as a result of their lower body weight relative to adults. The HI values were greater than 1 for both seasons for children, which indicates adverse non-carcinogenic risks. For adults, the HI values exceeded 1 in all but one location in the dry season, which also indicates adverse non-cancer risks of using this water for consumption purposes. For adults, the risk of developing adverse non-carcinogenic effects was reduced in the wet season.

Table 11 Hazard index and total cancer risk values for exposure of adults and children to water from Bomadi Creek
Full size table

In the case of carcinogenic risks, the computed values indicate a high risk of developing cancer in children and adults across both seasons, with all values exceeding the 1 × 10–6 United States Environmental Protection Agency (USEPA, 2010) permissible limit. This also corroborates with the water quality indices calculated in this study, which all indicate that the water from Bomadi Creek is not suitable for drinking and other domestic purposes.

Conclusion

The results presented in this study highlight the general quality of water from Bomadi Creek in the Niger Delta of Nigeria. A comparison of the physicochemical water parameters with their respective permissible limits indicated that all but turbidity were within the allowed limits, and that seasonal variation did not significantly affect the quality of the determined parameters in most of the studied locations. In the case of the metal concentrations, Cd, Fe, and Pb exceeded their permissible limits in most of the locations, and the elevated levels of these metals contributed significantly to the overall quality of the water, as indicated by the computed values for the water quality index (WQI) and the comprehensive pollution index (CPI). The overall water quality, as indicated by the WQI values (> 100), showed that the water is unsuitable for drinking and other domestic purposes. Similarly, the CPI values indicated serious pollution of the water. The implication of the results of this study is that water from Bomadi Creek is unsuitable for drinking, and there is a probable non-cancer and cancer risk associated with human consumption (especially for children) of the water from this creek. Remediation strategies need to be urgently put in place for continued use of this water source.