Papillært cystadenoma lymphomatosum, almindeligvis kendt som Warthins tumor, er en godartet vækst i spytkirtlerne, som udvikler sig langsomt og sjældent forårsager alvorlige helbredsproblemer. Denne ikke-kræftagtige tumor opstår typisk nær øret, hvor de største spytkirtler befinder sig, og rammer hovedsageligt ældre voksne, særligt dem der ryger.

Epidemiologi

Warthins tumor er den næsthyppigste godartede tumor, der påvirker spytkirtlerne, og den udgør cirka 2% til 15% af alle tumorer, der udvikler sig i parotiskirtlen. Parotiskirtlen er den største spytkirtel, placeret lige under og foran øret, og den er ansvarlig for at producere spyt, som hjælper med fordøjelsen og holder munden fugtig.^[3][4]

Denne tilstand rammer primært ældre personer, med højeste forekomst i specifikke aldersintervaller afhængigt af køn. Hos kvinder forekommer den højeste forekomst i deres sjette årti af livet, hvilket betyder mellem 50 og 60 år, mens hos mænd skifter den højeste forekomst til det syvende årti, mellem 60 og 70 år.^[3]

Historisk set blev mænd oftere ramt af Warthins tumor end kvinder, hvilket viste en klar mandlig overvægt. Imidlertid viser nyere medicinske rapporter, at denne kønsforskel er faldet over tid. Det ændrede mønster kan afspejle ændringer i rygevaner på tværs af forskellige befolkningsgrupper, da rygning er stærkt forbundet med udviklingen af denne tumor.^[3]

Tilstanden forekommer lejlighedsvis hos yngre patienter, selvom dette er mindre almindeligt. Det, der gør Warthins tumor særligt bemærkelsesværdig, er dens tendens til at påvirke begge parotiskirtler hos den samme person. Mellem 5% og 14% af tilfældene involverer bilateral præsentation, hvilket betyder, at tumorer udvikler sig på begge sider af ansigtet, selvom det typisk sker på forskellige tidspunkter snarere end samtidigt.^[1][4]

Årsager

Den præcise oprindelse og årsagerne til Warthins tumor er stadig genstand for løbende medicinsk debat, på trods af talrige teorier, der er blevet foreslået gennem årene. Nogle forskere antyder, at denne godartede vækst opstår på grund af visse tumorfremkaldende virkninger på epitelceller, som bliver fanget inden i tilstødende lymfeknuder eller i selve parotiskirtlen. Epitelceller er de celler, der danner det ydre lag af forskellige væv og organer, herunder kirtlernes slimhinde.^[3]

Flere faktorer er blevet identificeret som muligvis bidragende til udviklingen af Warthins tumor. Disse omfatter infektion med Epstein Barr-virus, en almindelig virus, der kan forblive sovende i kroppen, tobaksforbrug, autoimmune sygdomme hvor kroppens immunsystem angriber sit eget væv, eksponering for ioniserende stråling og kronisk inflammation, der fortsætter over længere perioder.^[3]

På trods af disse identificerede sammenhænge forbliver mange spørgsmål ubesvarede. Medicinske eksperter fortsætter med at undre sig over, hvorfor Warthins tumor overvejende påvirker mænd, og hvorfor tobaksforbrug specifikt påvirker parotiskirtlerne frem for de mindre spytkirtler, der befinder sig inde i munden. Disse forvirrende aspekter tyder på, at det komplette billede af, hvordan denne tumor udvikler sig, endnu ikke er fuldt forstået.^[3]

Risikofaktorer

Warthins tumor skiller sig ud som den eneste godartede spytkirteltumor med en stærk, veldokumenteret sammenhæng med cigaretrygning. Denne forbindelse er så betydningsfuld, at den definerer meget af det, vi ved om, hvem der udvikler denne tilstand.^[3]

Personer, der ryger cigaretter, står over for dramatisk højere odds for at udvikle Warthins tumor sammenlignet med den generelle befolkning. Rygere har cirka otte gange større risiko end ikke-rygere, hvilket gør tobaksforbrug til den vigtigste risikofaktor, der klart er blevet identificeret for denne tilstand. Denne stærke sammenhæng tyder på, at kemikalier fra tobaksrøg direkte kan påvirke udviklingen af disse tumorer i spytkirtlerne.^[1]

Alder repræsenterer en anden betydelig risikofaktor. Tumoren rammer sjældent unge mennesker og viser i stedet en klar præference for ældre voksne. Risikoen stiger væsentligt efter 50 år, hvor de fleste tilfælde diagnosticeres hos personer mellem 60’erne og 70’erne. Dette aldersrelaterede mønster tyder på, at enten langvarig eksponering for risikofaktorer eller naturlige aldringsprocesser i spytkirtlerne kan bidrage til tumorudvikling.^[1]

Selvom den mandlige overvægt, der blev observeret i tidligere årtier, er faldende, viste historiske data konsekvent mænd som værende i højere risiko end kvinder. Dette ændrede mønster kan afspejle udviklende rygevaner på tværs af kønnene, da flere kvinder begyndte at ryge i senere generationer, og flere mænd er stoppet med at ryge i de seneste årtier.^[3]

Symptomer

Warthins tumor præsenterer sig typisk på en måde, der forårsager minimal ubehag for patienterne, hvilket nogle gange kan føre til forsinket lægehjælp. Tumoren udvikler sig karakteristisk langsomt over måneder eller år og bliver gradvist større uden at forårsage smerte. Denne smertefri karakter adskiller den fra mange andre tilstande, der påvirker spytkirtlerne.^[1][2]

Det mest almindelige og mærkbare symptom er forekomsten af en masse eller hævelse nær vinklen af mandiblen, som er kæbebenets hjørne, hvor det vender opad mod øret. Denne placering svarer til halen af parotiskirtlen, det område hvor Warthins tumor oftest udvikler sig. Patienter bemærker typisk en knude, der føles blød eller fluktuerende ved berøring, hvilket betyder, at den kan føles som om den indeholder væske frem for at være fast.^[1][2]

Den fluktuerende kvalitet opstår på grund af tumorens indre struktur, som indeholder talrige cystiske rum fyldt med væske. Denne karakteristiske tekstur kan hjælpe med at skelne Warthins tumor fra andre typer af spytkirtelvækster under fysisk undersøgelse.^[2]

Fordi tumoren vokser langsomt og ikke forårsager smerte, kan patienter leve med den i mange år, før de søger lægehjælp. Et dokumenteret tilfælde beskrev en patient, som havde en asymptomatisk tumor i otte år, før vedkommende søgte behandling. Fraværet af presserende symptomer betyder, at mange mennesker kun konsulterer en læge, når massen bliver stor nok til at forårsage kosmetiske bekymringer, eller når de bliver bekymrede over dens tilstedeværelse.^[2][7]

I tilfælde, hvor tumoren påvirker begge parotiskirtler, kan patienter bemærke hævelse på begge sider af ansigtet, dog typisk på forskellige tidspunkter. Denne bilaterale præsentation kan indledningsvis forårsage bekymring, men den forbliver karakteristisk for denne godartede tilstand frem for at indikere noget mere alvorligt.^[1]

Forebyggelse

I betragtning af den stærke sammenhæng mellem Warthins tumor og cigaretrygning er den mest effektive forebyggende foranstaltning at undgå tobaksforbrug helt eller stoppe, hvis du i øjeblikket ryger. Da rygere står over for otte gange risikoen hos ikke-rygere for at udvikle denne tumor, reducerer eliminering af tobakseksponering markant sandsynligheden for tumorudvikling.^[1]

For mennesker, der ryger, kan rygestop-programmer, der omfatter rådgivning, adfærdsstøtte og potentielt medicin, hjælpe med at bryde afhængigheden af tobak. Fordelene ved at stoppe strækker sig langt ud over at reducere risikoen for Warthins tumor, da rygestop sænker risikoen for adskillige andre alvorlige helbredstilstande, herunder hjertesygdomme, slagtilfælde og forskellige kræftformer.^[1]

Fordi tumorens nøjagtige årsager forbliver ufuldstændigt forstået, og andre risikofaktorer som Epstein Barr-virusinfektion eller kronisk inflammation er vanskelige at kontrollere, er der ingen andre specifikke forebyggende strategier, der er vist sig at være effektive. Imidlertid kan opretholdelse af generelt godt helbred gennem regelmæssige lægetjek hjælpe med tidlig opdagelse, hvis en tumor udvikler sig.^[3]

Folk, der bemærker usædvanlig hævelse eller knuder nær deres kæbe eller under deres ører, bør søge lægeundersøgelse frem for at vente på, at symptomerne forværres. Tidlig opdagelse, selvom det ikke er forebyggelse i sig selv, giver mulighed for rettidig behandling, før tumoren vokser større, hvilket kan gøre kirurgisk fjernelse simplere og reducere risikoen for komplikationer.^[7]

Patofysiologi

Forståelsen af, hvad der sker inde i Warthins tumor på mikroskopisk niveau, hjælper med at forklare dens karakteristika og adfærd. Tumoren har et karakteristisk udseende under mikroskopet, som patologer let kan identificere, hvilket gør diagnosen mere ligetil, når vævsprNext Article in Journal
Assessing Spatial and Temporal Variability of Deep Drainage and Runoff in the Context of Irrigation Management
Next Article in Special Issue
Temporal Trends and Their Spatial Patterns in Water Quality in the Mun River Basin, Thailand
Previous Article in Journal
Ecological Effects of a Sewage Treatment Plant on a Densely Populated Mediterranean Coastal System (Savona, NW Italy)
Previous Article in Special Issue
Spatiotemporal Variation and Driving Factors of Water Quality in Danjiangkou Reservoir under the Influence of South-To-North Water Diversion: A Five-Year Analysis from 2016 to 2020

Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Estimation and Trend Detection of Water Quality in Ile-Ife, Nigeria Using Multivariate Statistical Techniques

1
Department of Geography, Obafemi Awolowo University, Ile-Ife 220005, Nigeria
2
Department of Public Health and Health Sciences, University of Parma, 43121 Parma, Italy
3
Department of Earth and Environmental Studies, Montclair State University, Montclair, NJ 07043, USA
4
Geology Department, School of Environment & Earth Sciences, Central University of Punjab, VPO-Ghudda, Bathinda 151401, Punjab, India
*
Authors to whom correspondence should be addressed.
Water 2023, 15(2), 274; https://doi.org/10.3390/w15020274
Submission received: 22 November 2022 / Revised: 5 January 2023 / Accepted: 6 January 2023 / Published: 9 January 2023
(This article belongs to the Special Issue Impacts of Land Use Change on Water Quality and Ecosystem Services)

Abstract

:
This paper assessed the trends and qualities of water from sources including hand-dug wells and boreholes at residential, commercial, and industrial areas of Ile-Ife Local Government, Nigeria. Forty-three water quality parameters were analyzed, including physicochemical and bacteriological parameters at residential, commercial, and industrial areas during both dry and rainy seasons. The trend analysis, principal component analysis (PCA), and Mann–Kendall test were used to establish the relationships among water parameters, and Piper plot was used to determine the hydrochemical facies of water sources. Results of the Mann–Kendall test indicated a significant (p < 0.05) temporal increasing trend in the levels of Ca2+, Mg2+, K+, NH4+, S2−, SO42−, Cl, PO43−, COD, NO3, NO2, Mn, Zn, Pb, Cd, Cr, Fe and E. coli. The result of the PCA indicated that the first four principal components (PC) accounted for 85.97% and 81.60% variance in the dry and rainy seasons, respectively. The water quality variables of land use types were significantly different (p 0.05 values were assumed to be normal and data with significant values p < 0.05 were assumed to be not normal. Also, the Mann–Whitney U-Test (a non-parametric test), was applied for two independent samples analysis of any pair of variables from residential, commercial, and industrial land uses. The p-values of <0.05 were considered statistically significant at a 95% level. Mann–Kendall significance test and Sen's Slope were used for analyses of the trend of time series for water quality variables and the magnitude of their trends, respectively [53,54,55]. The PCA for Windows 4.1 was employed to perform PCA and Scree plots of water quality variables at the study area. In addition, a Piper plot was employed to determine the hydrochemical facies of water samples of hand-dug wells and borehole of the study area [51].

2.5. Normality Test

The Kolgomogorv–Smirnov (K-S) normality test is a non-parametric technique that evaluates the differences or compares the distributions of a sample to find out if the sample comes from a population with a normal distribution. According to Ghasemi and Zahediasl [56], the K-S is an important statistical tool to normalize skewed data sets that can be applied before performing a certain statistical test to observe the relationship between variables. The K-S test compares the cumulative distribution of two data sets, one observed and the other theoretical. Then it computes a p-value based on this. The null hypothesis for this test is that the observed data follow the specified distributions. If p-values are less than the 0.05 level of significance, then it is rejected and assumed the data do not come from the normal distribution [57]. However, if the p-values are higher than 0.05, then the observed data follow the normal distribution (H0 is accepted). For a reliable result, the data were transformed using log10 values. However, Ca2+, Mg2+, Na+, K+, Cl, S2−, PO43−, COD, BOD, Mn, Zn, Pb, Cd, Cr, Fe, Ni, and TC were all significantly not normally distributed (K-S. p 0.05), which were temperature, turbidity, DO, CO32−, HCO3, NH4+, NO3, and NO2. Therefore, with data that are not normally distributed, the non-parametric test (Mann–Whitney U) was applied for comparisons of the mean of water quality variables among the three land uses [56,58].

2.6. Principal Component Analysis (PCA)

PCA is a widely used technique in various environmental and water quality studies around the world [37,59,60,61]. The PCA technique is a multivariate approach that reduces the complexity of a large number of data sets by explaining only the variances of the data. Therefore, it removes high correlations among the large number of original variables and transforms them into independent linearly uncorrelated new variables (called principal components (PCs)), with minimum loss of information at the original data set [37,62,63]. The PCA basically explains the correlation or covariance structure of the data set. Thus, the first principal component (PC1) is likely to display the highest variation in the data set, and the second PC (PC2) represents the second-highest variation in the entire data set and so on, in this order until the total variance that exists in the data is explained [64]. To determine the spatial and temporal water quality trends, the varimax rotation technique was used to minimize the effects of weather and scale with Kaiser Normalization [14]. In this study, PCA was applied and the varimax rotated component matrix was calculated using 43 parameters at residential, commercial, and industrial areas. The PCA techniques were performed by the PCA for Windows 4.1, which calculated variances in the data sets and rotated varimax loading factors. The Scree plot was employed to determine the number of PCs which could help explain the variances in the data sets. However, eigenvalues and total variance explained by each PC were computed and only PCs with eigenvalues > 1 were considered for data interpretation [31,32,64]. The loading scores were classified as high, moderate, and weak. According to Liu, Lin, and Kuo [31], PCA greater than 0.75 are considered strong, those between 0.75 and 0.50 are moderate, and those within 0.50 and 0.30 are weak.

2.7. Mann–Kendall Trend Analysis

The Mann–Kendall test (M-K) is a popular non-parametric trend test that is used extensively around the world, to quantify temporal trends in series data [39,65]. In this study, the analyses were carried out as follows: first, a non-parametric Theil-Sen Slope (TSS) line was used to estimate the rate of change in trends (p-values of i.
The H statistic of the Mann–Kendall was tested using Equations (3) and (4) [66], as follows:
H = i = 1 N 1 j = i + 1 N sgn ( X j X i )
sgn ( X j X i ) { 1 , if ( X j X i ) > 0 0 , if ( X j X i ) = 0 1 , if ( X j X i ) 0) or decreasing (H < 0) trend in the series. In a zero (0) trend, H is approximately normally distributed [39,67], provided that the following assumptions are valid: the distribution of all the Xi, is mutually independent; and the distribution of all Xi is identical. To compute the mean and variance of H, the non-parametric approach [66,67] for trend analysis was used. The summary of the Mann–Kendall trend technique as well as Sen's Slope magnitude equation was computed and presented in Table 1.

2.8. Piper Plot

The Piper plot was used to understand the hydrochemistry of water resources [5,51,68]. According to Belkhiri, Boudoukha, Mouni, and Baouz [69], the Piper is made up of two triangles (one for cations and another for anions) and one diamond plot. Thus, the points from the two triangles are extrapolated into the central diamond plot [69]. The Piper diagram shows the concentrations of cations and anions as a percentage of milliequivalents. The Piper plot does not show total dissolved solids. Therefore, the Piper plot was plotted using XLSTAT and Microsoft Office 2016 to show six distinct hydrochemical facies. These facies include (1) CaHCO3 type, (2) NaCl type, (3) Mixed Ca-Na-HCO3 type, (4) Mixed CaMgCl type, (5) CaCl type, and (6) NaHCO3 type.

2.9. Overall Groundwater Pollution Index

The overall groundwater pollution index (GPI), was applied to understand the spatial variation and extent of contamination in various kinds of water bodies (groundwater, surface water, and drinking water) [70,71]. The GPI technique is appropriate to assess groundwater pollution. This method examines the suitability of groundwater for drinking purposes in water quality assessments [70]. The GPI is calculated using Equations (5)–(8) [70,71,72]. This technique was applied to analyze 26 groundwater samples collected during the rainy season. According to Karunanidhi, et al. [73], groundwater is more susceptible to contamination during the rainy season due to infiltration. The results of GPI are presented in five classes, in which S
w = k A V
where AV is the assigned value. The estimated GPI was classified into five classes (Table S2) [73,74,75].

3. Results

3.1. Descriptive Statistics

Descriptive statistics were employed to show the minimum, maximum, mean, and standard deviation of water quality parameters from residential, commercial, and industrial areas of the study location in both dry and rainy seasons. Kurtosis and skewness were calculated to show the normality of the data sets. Summary statistics in the dry season at residential, commercial, and industrial are shown in Table 2 whereas those of the rainy season are presented in Table 3.
Generally, the results in the dry season (Table 2) showed the average temperature ranges from 27.9 ± 0.87 to 28.5 ± 0.91 °C; pH ranges from 7.13 ± 0.37 to 7.34 ± 0.27; TDS ranges from 189 ± 74.7 to 236.4 ± 113.2 mg/L; EC values range from 311.0 ± 109.9 to 383.9 ± 184.7 µS/cm; turbidity ranges from 1.54 ± 1.01 to 2.35 ± 1.93 NTU; TH ranges from 105.4 ± 33.94 to 110.8 ± 38.74 mg/L; Mg ranges from 13.99 ± 5.51 to 17.60 ± 6.13 mg/L; Ca2+ ranges from 28.41 ± 7.77 to 36.73 ± 21.57 mg/L; DO ranges from 5.30 ± 0.99 to 6.14 ± 1.51 mg/L, whereas Na ranges from 28.50 ± 11.11 to 34.95 ± 14.65 mg/L. The average level of S2− ranges from 0.04 ± 0.02 to 0.04 ± 0.02 mg/L; CO32− ranges from 0.08 ± 0.09 to 0.13 ± 0.17; and PO43− ranges from 0.01 ± 0.01 to 0.03 ± 0.02 mg/L. The concentration of trace metals varies from one location to another. For example, Mn ranges from 0.13 ± 0.11 to 0.19 ± 0.16 mg/L; Pb ranges from 0.09 ± 0.05 to 0.01 ± 0.01 mg/L, whereas Cd ranges from 0.011 ± 0.01 to 0.01 ± 0.01 mg/L. Bacteriological organisms, including TC and E. coli indicated pollution in groundwater of the study area. Concentrations of E. coli range from 0.25 ± 0.45 to 0.80 ± 0.90 mg/L.
Descriptive statistics of water variables in both residential, commercial, and industrial land uses during the rainy season are presented in Table 3. Most of the water quality variables exceeded the permissible limits, particularly for Cl, Pb, Fe, TC, and E. coli. In comparison between the concentrations of variables in the dry season, the levels of water quality parameters increased appreciably. For example, the concentration of water temperature ranges from 28.0 ± 1.20 to 28.9 ± 0.49 °C; pH ranges from 7.07 ± 0.39 to 7.71 ± 0.34; TDS ranges from 253.1 ± 108.7 to 379.4 ± 178.9 mg/L; TH ranges from 142.8 ± 34.37 to 169.0 ± 59.7 mg/L; turbidity ranges from 2.13 ± 1.28 to 3.62 ± 1.49 mg/L; DO ranges from 5.58 ± 1.11 to 6.06 ± 1.22 mg/L; Ca2+ ranges from 42.13 ± 13.85 to 54.40 ± 28.62 mg/L; Na+ ranges from 38.13 ± 15.83 to 47.27 ± 19.69 mg/L; S2− ranges from 0.065 ± 0.06 to 0.09 ± 0.08 mg/L; PO43− ranges from 0.039 ± 0.02 to 0.053 ± 0.05 mg/L, whereas Mn ranges from 0.24 ± 0.18 to 0.33 ± 0.24 mg/L. The concentrations of Pb, Cr, Ni, Cd, and Fe were higher during this season in all locations. Bacteriological parameters (TC and E. coli) were higher and exceeded the WHO [46] permissible limits at all the locations, for example, TC ranged from 2.07 ± 1.22 to 2.33 ± 1.50 mg/L, whereas E. coli ranged from 0.75 ± 0.84 to 1.6 ± 0.91 mg/L. The Mann–Whitney test was performed to compare the means of the measured variables collected from the three locations in dry and rainy seasons. The results indicated significant differences (H = 0.05) among the land use types (residential, commercial, and industrial) for 30 variables including Temp, EC, Turbidity, TDS, TH, pH, Mg2+, Ca2+, Na+, NH4+, S2−, PO43−, SO42−, Cl, COD, BOD, NO3, NO2, Mn, Zn, Pb, Cd, Cr, Fe, and E. coli. The results imply that the effect of land use change in the study area has contributed to differences in the water quality for all land use types.
For comparative and easy interpretation of the result, this study applied correlation analysis to show the level of relationships between all the measured water variables and all the land uses (residential, commercial, and industrial). Therefore, the result of Pearson’s correlation analysis is presented in (Supplementary Tables S3 and S4). The correlation results in (Supplementary Tables S3 and S4) revealed that variables TDS, EC, TH, Ca2+, SO42−, BOD, and NO2 had significant (p 1, which explained 85.97% of the total variance in the dataset (Table 4). The Scree plot was used to determine the importance of PCs (Figure 2a).
The first principal component (PC1) in the dry season at the study location has the highest value of eigenvalue with 17.04, and this accounted for 47.33% of the total variance in the entire data set. In PC1, 18 parameters showed a high loading factor (≥0.75) with strong positive correlations. These consist of TDS, EC, TH, turbidity, Ca2+, Mg2+, Na+, NO2, F, BOD, SO42−, Cl, Mn, Zn, Fe, E. coli, TC, and Pb. Both EC and TDS had the highest correlations with the value of 0.976 and 0.956, respectively. All these variables had strong correlations with a few outliers. Therefore, PC1 could be called TDS, EC, and TH dominated components, and also represent water containing cations and anions of solutes. Thus, the high loading of EC, TDS, TH, Mn, Zn, Fe, and SO42− in the dry season indicates anthropogenic pollution from domestic sewage, agrochemicals, and domestic discharge, with some of these parameters originating from natural sources as well. In addition, the result of the descriptive statistics of PC1 indicated that both residential and commercial areas have high values of the abovementioned parameters. Thus, the outcome of the positive correlation in PC1 is due to the high values of EC, TDS, TH, SO42−, Mn, Fe, Ca2+, Mg2+, Zn, NO2, and E. coli at residential and commercial locations. For example, the residential areas are where most residents dispose of solid wastes from homes onto the ground or discharge home effluents (sewage and refuse water) directly onto the ground without treatment.
In contrast, there is only one variable in PC2 that showed a high loading factor (≥0.75), namely pH, with eigenvalue = 6.95 and explained 19.31% of the total variance. The parameters of PC2 had a strong negative correlation, which implies that the concentrations of the parameters were not high during the dry season and at the study locations. This indicates that pH was moderately alkaline in the study locations. Hence, PC2 can be termed the moderately alkaline pH component.
The third PC (PC3) has an eigenvalue of 4.62 and explained 12.84% of the total variance with ten parameters showing moderate to weak loading factors (PO43−, NO3, Cd, Cr, and Ni), which is between 0.50 and 0.75. The positive correlations of the loading factor of these water variables is possibly attributed to anthropogenic pollution from leakages of underground storage tanks of petroleum products into groundwater, dumping of wastes, effluents from industries, agricultural fertilizers, and herbicides, and geological influences. The variability of PC3 can be termed the anthropogenically induced PO43−, and NO3 component because PO43− and NO3 showed strong correlation in PC3 followed by Cd, Cr, and Ni.
However, PC4 has an eigenvalue of 2.37 and explained 6.59% of the total variance with four parameters showing moderate to weak loading (CO32−, HCO3, DO, and K+) ranging between 0.37 and 0.75. The results showed a moderate correlation between variables DO and K+, while a weak correlation was observed between variables CO32− and HCO3. These sources of variance are attributed to natural processes. Thus, the variability of PC4 is simply referred to as DO and K+ components.

3.3.2. PCA for the Rainy Season

The result of PCA analysis during the rainy season (Table 4) indicated the extraction of four principal components with eigenvalues > 1. These had accounted for 81.60% of the total variance in the data sets. The Scree plot was used to determine the importance of PCs (Figure 2b). The first principal component (PC1) has eigenvalue = 15.53, and explained 43.14% of the total variance with 20 variables having strong loading factors >0.75, including EC, TDS, TH, turbidity, Ca2+, Mg2+, Na+, K+, NH4+, S2−, SO42−, Cl, PO43−, NO3, COD, Mn, Pb, Cd, Fe, and E. coli. These had accounted for the highest value of total variance among all the extracted PCs. The high value of loading factors of these variables in PC1 is most likely due to their high values in residential and commercial areas of the study location, which showed a strong positive correlation among all the parameters (Supplementary Table S4) compared with a weak negative correlation with pH (Table 4). Thus, the variability in PC1 could be referred to as TDS, EC, and TH components similar to the dry season, which indicate water containing cations and anions of solutes. These are found in the residential and commercial area where the concentration of these variables was higher during the rainy season.
The second principal component (PC2) accounted for 20.05% of the total variance with eigenvalue 7.22, and has seven parameters with loadings factors >0.75 including HCO3, F, COD, BOD, Cr, Ni, and TC. The high loading of COD and BOD indicated organic pollution, which may be attributed to anthropogenic influences from sewage, discharge from industries, and agrochemicals. In addition, high correlation among these variables shows a common source of contaminants, which is probably from industries and sewage of residential and commercial areas, with effluents discharged directly into the ground. Thus, the variability of PC2 can be called industrially induced COD-BOD and TC components because of the direct influence of industries at the study area. The third principal component (PC3) accounted for 11.88% of total variance with an eigenvalue of 4.28. The PC3 has one moderate and one weak loading factor with CO32− and pH. These have a negative weak correlation with other water variables such as K+, NO3, and Pb, which showed that they were not high at the study locations. Finally, the fourth principal component (PC4) has eigenvalue = 2.40 and contributed 6.53% of the total variance, which showed only two variables with moderate loading factors ≥0.75, namely Zn and NO2. Notably, there was a weak positive correlation between Zn and K+, which is probably a result of anthropogenic pollution from the metal industry at the commercial and industrial areas, automobile servicing and repair workshops, and runoff from roads and domestic discharge. The variability of PC4 can be called a Zn and NO2 component.
Generally, in both dry and rainy seasons, four principal components (PCs) were extracted using the Scree plot; thus, PC1 exhibited most of the variance in the entire data sets. This indicates that PC1 showed the contribution of each parameter in the overall data set. The variables that most contributed to PC1 in both seasons were EC, TDS, TH, turbidity, Ca2+, Mg2+, Mn, Zn, and Fe, which were attributed to increasing anthropogenic activities and natural factors (weathering and leaching of chemical elements from host rocks and soils into groundwater). In the study, PC2, PC3, and PC4 explained much smaller variances; thus, parameters were easily and mostly attributed to anthropogenic pollution (runoff with high concentration of nutrients and pollutants, sewage, effluents from industries, dumping of wastes from residential and commercial activities, construction works, and application of agrochemicals on agricultural fields, automobile servicing, and repairs, etc.) and from natural (weathering of rocks, leaching and lixiviation) sources. Generally, PC3 and PC4 explained little variance in the variances in the entire data sets. Particularly PC4, which was characterized by weak to moderate loadings of pH, CO32−, HCO3, DO, and K+ in the dry season, and by pH, CO32−, Zn, and NO2 in the rainy season, is most probably from natural sources from weathering and ions exchange. According to Liu, et al. [31], the increase of pH is caused by a high concentration of bases such as K+. This justifies that K+ is being contributed to groundwater through cation exchange [77]. The percentage of total variance explained increased in the dry season compared with the rainy season. However, the number of parameters contributing to PC1 during the rainy season increased. This is because there is high precipitation in the rainy season, which leached various solutes, pollutants, and organic and inorganic particles from roads, wastes, and landmass, as well as soaking them away and eventually percolating into hand-dug wells and boreholes, thus contributing a high volume of contaminants and pollutants into water sources. These results correspond to the findings of Banda, et al. [77].

3.4. Hydrochemistry and Piper Plot

Generally, the chemistry of groundwater determines the sources of dissolved compounds or ions in hand-dug wells and boreholes, which could be due to the host rock materials of groundwater aquifers and the effect of human activities at the ground surface. The Piper trilinear diagram was used to determine the hydrochemical composition of samples at the study areas. Six hydrochemical facies (Ca–HCO3, Na-Cl, Mixed Ca-Na-HCO3, Mixed Ca-Mg-Cl, Ca-Cl, and Na-HCO3 types) were determined using the Piper plot. The Piper plot of residential areas during the dry season (Figure 3a) showed most of the residential samples were classified as Ca–HCO3 type (80% of the samples). This indicates an alkaline earth exceeding alkalis, and weak acids exceeding strong acids, and represents high levels of Calcium and Carbonate ions. Therefore, this hydrochemical facies is associated with a recharge zone and shallow aquifer which characterized an early stage of geochemical evolution, resulting from silicate weathering [5,69].
However, 40% of the commercial samples indicated Ca–HCO3 type and the remaining 60% is classified under the Ca-Mg-SO4 type. Thus, this hydrochemical facies can be termed as alkaline earth exceeds alkalis, and strong acids exceed weak acids, and are characterized by higher levels of Ca and SO42−. Additionally, this hydro-facies indicates non-carbonate hard water. This is associated with the discharge zone and the deep aquifer found within an intermediate stage of geochemical evolution [5,69]. However, samples from industrial areas (50%) belong to the Ca-Mg-SO4 types; the remaining 50% of the samples belong to the Ca–HCO3 type, which is due to the geology of the study area and anthropogenic pollution from industrial activities. The Piper plot for the rainy season indicated that 66% of residential groundwater samples belong to the Ca–HCO3 type while 34% of samples belong to the Ca-Mg-SO4 type (Figure 3b), which indicates alkaline earth exceeds alkalis, and strong acids exceed weak acids, and are characterized by higher levels of Ca and SO42−. This indicated non-carbonate hard water associated with shallow and deep aquifers within an intermediate stage of geochemical evolution [5,69]. However, more than half of the samples (66%) from the commercial area belonged to the Ca-Mg-SO4 type while 34% of samples were under the Ca–HCO3 type. This indicates that in a higher number of locations within commercial land use, there was alkaline earth exceeding alkalis, and strong acids exceeding weak acids, and were characterized by higher levels of Ca and SO42−. Thus, this hydro-facies is called Ca-Mg-SO4 type, which is associated with the discharge zone and the deep aquifer found within an intermediate stage of geochemical evolution [5,69]. Samples from industrial areas (33%) belonged to the Ca–HCO3 type, whereas 67% belong to the Ca-Mg-SO4 type; thus, more of the samples in industrial areas belonged to non-carbonate hard water associated with the discharge zone and deep aquifer within an intermediate stage of geochemical evolution [5,69]. Therefore, hydrogeochemically, groundwater samples from three land uses (residential, commercial, and industrial land use) in the study area are characterized by calcium, magnesium, carbonate, and sulfate ions, hence termed Ca-Mg-HCO3-SO4 type. As stated by Gomo and Van Tonder [78], this hydrogeochemistry (Ca-Mg-HCO3-SO4 type) is associated with rapid recharge processes and/or short residence times of groundwater in an aquifer system. In view of that, the types of rocks at the study location are composed of migmatite gneiss, granite gneiss, coarse-grained granite, and pegmatite.

3.5. Overall Groundwater Pollution Index (GPI)

The overall groundwater pollution index (GPI) of the rainy season samples from residential, commercial, and industrial areas were computed in which total dissolved solid, turbidity, pH, DO, calcium, magnesium, potassium, bicarbonate, carbonate, nitrate, nitrite, sulfate, chloride, phosphate, fluoride, manganese, zinc, iron, E. coli, and total coliform were considered as computed parameters. The overall GPI results presented an insignificant to low-moderate pollution rate in the rainy season in all the samples that were analyzed. Spatial interpolation technique of inverse distance weight (IDW) was used to interpolate the overall GPI map of the study location in Figure 4. Inverse distance weight (IDW) was employed because of its simplicity and high accuracy in spatial interpolation. For example, seven (28%) samples were classified as insignificant, seventeen (68%) samples were classified as low-moderate contamination, and only one (4%) sample was classified as high overall pollution and located at industrial area (GPI = 157) as shown in Figure 4.
Generally, a high proportion of samples (68%) in the study area had low-moderate overall pollution. The results suggest that water sources (hand-dug wells and boreholes) were polluted during the rainy season, which required urgent treatment before human consumption to avert waterborne diseases and potential health risks. These sources of pollution can be attributed to sewage, effluents from industries, dumping of wastes at open places, discharge of run-off, and fertilizers and chemicals in agricultural areas.

4. Discussion

4.1. Factors Controlling the Spatial Variations in Water Quality

The PCA technique revealed that the highest variances of measured water quality were explained in the dry and rainy seasons by 85.97% and 81.60%, respectively. In both the dry and rainy seasons, PC1, accounted for the highest variance, at 47.33% and 43.14%, respectively. The remaining variance (53.67 and 56.84%) in the dry and rainy seasons, were caused by other factors not analyzed in this study. Therefore, the results of the PCA could help to understand the factors and causes responsible for the variability in water quality from different locations (residential, commercial, and industrial areas) in both seasons. For example, PC1 represented the highest variance in both the dry and rainy seasons, having strong loadings of TDS, EC, TH, turbidity, Ca2+, Mg2+, Na+, NO2, F, BOD, SO42−, Cl, Mn, Zn, Fe, E. coli, TC, and Pb. The high concentration of turbidity in PC1 is associated with a high level of E. coli which explained the pollution from sewage, poor sanitary conditions of hand-dug wells and boreholes, and leachates that originated from domestic discharge, wastes, and run-off, particularly in the rainy season. Higher turbidity in groundwater means high dissolution of solutes, which is the cause of higher values of TDS, EC, TH, and anions. High turbidity in PC1 was associated with shallow and uncapped hand-dug wells and boreholes, particularly in residential and commercial areas, which led to direct surface run-off into the shallow hand-dug wells during rainfall periods. Furthermore, higher concentrations of Mn, Zn, Fe, and Pb in the PC1 during the dry and rainy seasons in residential and commercial areas could have originated from anthropogenic sources such as sewage, discharge of effluents from the dumping of wastes, farming activities, and auto-mobile workshops and servicing centers, with the addition of natural sources from host rocks and soils. However, high concentrations of these metals could pose long-term potential health risks. Therefore, results of PC1 revealed variations in the dry and rainy seasons, this may be attributed to increasing urbanization and anthropogenic activities, which has caused an excessive increase of solutes in the groundwater during the wet season [18,79].
The high loading of Zn, and NO2 in the rainy season is attributed to effluents from industries, batteries, and bulb industries and automobile servicing workshops in the commercial area. Moreover, the increase of variables such as BOD, COD, and TC in the rainy season is attributed to effluents from industries, sewage, and wastes dumped from residential and commercial areas. Thus, PC2 in the rainy season is classified as industrially induced COD-BOD and TC components. In addition, moderate to strong loading of PO43− and NO3 in PC3 of the dry season indicated pollution from sewage, agricultural fertilizers, and dumping of wastes such as kitchen-waste, and animal droppings which might have infiltrated into groundwater [76,80,81]. PO43− does not exist naturally in groundwater; thus, the existence of PO43− in groundwater suggested its anthropogenic origin from sewage, industries, and agrochemicals [82,83,84]. Generally, in this study, both human activities and natural sources were contributing factors that altered the groundwater quality. However, the effect of human activities was higher than natural sources, particularly in the rainy season.
The major cations (Ca2+, Mg2+, K+, and Na+) and major anions (CO32−, HCO3, and SO42−) could have originated from both host rocks weathering and anthropogenic sources. These results support the study of Karunanidhi, Aravinthasamy, Subramani, Muthusankar, Rajkumar, and Sajil Kumar [31], who reported the influence of natural and anthropogenic sources on water quality. For example, Kodom [85], stated that potassium was released by the weathering of clay minerals, feldspars, and clays. Appelo and Postma [86] revealed that Ca2+, Mg2+, Na+, K+, and HCO3 could have been influenced by the oxidation of organic matter and minerals dissolution. Moreover, sulfate could have been attributed to weathering of rocks containing sulfide minerals, such as pyrite, and discharge of effluents from industries.

4.2. Trends and Temporal Variation in Water Quality Parameters

The results of temporal variation analysis of this study presented significant trends in 18 of the 43 parameters measured. The temporal variation that was observed in the parameters examined is attributed to an increase in temporal seasonal variations and anthropogenic activities in the study areas. The result of the Mann–Kendall test indicated that there was a significant (p 0.05) in both seasons. This is possibly caused by either seasonal or anthropogenic sources. However, the most significant trend that was observed included a significant increasing trend (p ≤ 0.05) in many parameters, which could be attributed to the increasing anthropogenic activities. For example, Pb, Cr, Zn, Cd, and Fe revealed a significant increasing trend (p ≤ 0.05), and exceeded the World Health Organization’s standard limits of 0.01, 0.05, 3.0, 0.003, and 0.3 mg/L, respectively, during the dry and rainy seasons at residential and commercial areas which may be attributed to runoff from roads, industrial effluents, sewage, leachates from dumpsites and waste materials, and automobile servicing workshops. In addition, there was a significant increasing trend in COD, E. coli, TC, and bacteriological parameters including total coliforms and E. coli, which exceeded the WHO [46] recommended standard of 0 colony forming unit (CFU/100 mL) of water in both seasons. This could be attributed to the influence of sewage, discharged effluents from residential and commercial areas, open defecation, and proximity of hand-dug wells and boreholes to sanitary facilities and dumpsites. The trend that was observed among all the measured parameters underlines the important and urgent need to implement appropriate and sustainable measures and policies to reduce the level of water pollution from groundwater in the study areas. Thus, the possible health risks associated with the exceedance of permissible limits of the measured parameters can be eliminated.

4.3. Spatial Variation of Water Quality and Pollution Sources

Comparison of three land uses (residential, commercial, and industrial areas) reveals variations between water quality parameters in both seasons. The variability of measured parameters as presented in Figure 5 revealed an increasing trend from residential to industrial areas with a few parameters that had a decreasing trend from residential to industrial areas; such parameters include temperature, turbidity, Mg2+, Na+, K+, BOD, NH4+, and S2−. However, more than half of the parameters showed an increasing trend from residential to industrial areas in both dry and rainy seasons. For example, parameters such as TDS, EC, DO, pH, Ca2+, Cl, HCO3, CO32−, SO42−, NO2, NO3, F, COD, Cr, Fe, Zn, Pb, Mn, Cd, Ni, and TC increased from residential to industrial areas. Comparing temporal variations of parameters in the dry and rainy seasons, most of the parameters increased during the rainy season, as presented in Figure 5 and Table 2 and Table 3. These patterns follow an increasing population, anthropogenic activities, and runoff, which was a major contributing factor in the rainy season. Thus, dissolved ions and organic particles in the runoff infiltrated into surface water and groundwater, and thereby contaminated it in all locations. However, organic pollution and trace metal pollution were observed in all locations. A slightly higher concentration was observed in samples from the residential and commercial areas compared with the industrial area.
According to WHO [46] classifications for suitability of groundwater for drinking purposes, the levels of pH, DO, TDS, EC, TH, turbidity, and major cations (Na+, K+, Mg2+, and Ca2+) were within the permissible limits in both seasons at all locations, particularly during the dry season. However, the majority of trace metals, bacteriological, and organic pollution indicators exhibited the levels that exceeded WHO [46] standards. All the measured trace metals including Mn, Zn, Pb, Cd, Cr, Fe, and Ni were higher in the residential and commercial areas in both seasons. For instance, at residential areas, the levels of Mn ranged from 0.13 ± 0.11 to 0.33 ± 0.17 mg/L; Zn ranged from 1.99 ± 1.30 to 3.40 ± 1.40 mg/L; Pb ranged from 0.26 ± 0.08 to 0.32 ± 0.10 mg/L; Cd ranged from 0.01 ± 0.01 to 0.02 ± 0.01 mg/L; Cr ranged from 0.02 ± 0.03 to 0.06 ± 0.03 mg/L; Fe ranged from 1.08 ± 0.74 to 1.54 ± 0.91 mg/L; Ni ranged from 0.01 ± 0.02 to 0.04 ± 0.04 mg/L; and COD ranged from 1.81 ± 0.65 to 2.45 ± 0.76 mg/L in dry and rainy seasons. Moreover, concentrations of trace metals and bacteriological parameters during the rainy season were higher than in the dry season. This is possibly because a large volume of contaminants, ions, and particles were washed out and carried along with the runoff on an impermeable paved road, then entered directly into hand-dug wells and boreholes through land surfaces and shallow uncapped wells, which increased the concentration of trace metals, bacteria, and organic pollution in water sources. The concentrations in the rainy season in this study correspond with the findings of Kumar, Bhat, and Rajan [70] and Karunanidhi, Aravinthasamy, Subramani, Muthusankar, Rajkumar, and Sajil Kumar [73] in India as well as Wagh, Mukate, Panaskar, Muley, Sahu, and Jadhav [75] in Nigeria who reported that groundwater is more susceptible to contamination in the rainy season due to infiltration of anthropogenic pollutants and runoff from agricultural fields, industrial areas, and sewages from residential areas into hand-dug wells. Therefore, there is a need for sustainable water management systems and effective water treatment for residential water users. These can be achieved through the following processes: (1) regular monitoring and assessment of water sources in order to know the levels of pollutants and contaminations; (2) point and non-point pollution sources should be brought under strict control to check excessive pollution of water sources; (3) contaminated water such as this should be treated before being supplied to residents to avoid potential hazard health risks; (4) there is a need for awareness, environmental education for all levels of society, and training of skilled personnel for water resources management to conserve groundwater; (5) frequent inspection and public awareness should be implemented along with strict adherence to government policies and regulations regarding environmental protection; (6) pollution of water sources from sewage, industrial effluent, discharge from residential and commercial areas, and dumping of wastes should be avoided to prevent organic pollution, trace metal pollution, and bacteriological contamination; (7) the construction of water harvesting structures and artificial recharge facilities could be an effective strategy for water resources management.

5. Conclusions

In this study, the quality and trend detection of water were estimated using 43 parameters from hand-dug wells and boreholes at residential, commercial, and industrial areas during the dry and rainy seasons. The results of descriptive statistics revealed that, in the dry season, the average levels of trace metals (Mn, Cd, Fe, Ni, Zn, Pb, and Cr) were higher compared with the rainy season. However, during the rainy season, both major cations (Ca2+, Mg2+, K+, and Na+) and anions (CO32−, HCO3, S2−, SO42−, Cl, NO2, NO3, PO43−) increased compared with the dry season. Organic pollution indicators such as BOD and COD as well as bacteriological parameters of TC and E. coli were higher during the rainy season.
The Mann–Kendall test indicated a significant increasing trend (p 0.05), SS: Significant Increasing trend (p ≤ 0.05).
Table 2. Descriptive statistics of water quality parameters at residential, commercial, and industrial areas during dry season.
Table 2. Descriptive statistics of water quality parameters at residential, commercial, and industrial areas during dry season.
Water VariablesResidential (n = 15)Commercial (n = 25)Industrial (n = 10)WHO 2017
MeanMinMaxSD± SEMeanMinMaxSD ± SEMeanMinMaxSD ± SE
Temp (°C)27.9 ± 0.8725.029.01.58 ± 0.2328.1 ± 0.6426.029.01.01 ± 0.2128.5 ± 0.9127.030.01.26 ± 0.37-
pH7.3 ± 0.276.68.00.36 ± 0.067.3 ± 0.356.97.80.56 ± 0.127.1 ± 0.366.87.60.49 ± 0.146.5–8.5
TDS189.0 ± 74.6782.0318.0124.43 ± 20.38236.4 ± 113.1593.0490.0175.16 ± 37.41207.2 ± 62.30146.0347.086.3 ± 25.03500.0
EC (µS/cm)311.0 ± 109.89139.0547.0183.17 ± 29.99383.9 ± 184.72151.0797.0285.18 ± 60.92339.1 ± 105.20250.0598.0145.5 ± 42.191250
Turbidity (NTU)2.3 ± 1.930.366.23.22 ± 0.531.7 ± 0.970.254.11.50 ± 0.321.5 ± 1.010.334.21.40 ± 0.415
TH105.4 ± 33.9458.8168.056.66 ± 9.28110.3 ± 41.3254.2179.763.86 ± 13.64110.8 ± 38.7453.4191.053.63 ± 15.55500
DO5.5 ± 1.133.57.41.89 ± 0.316.0 ± 1.341.99.32.06 ± 0.445.3 ± 0.993.66.71.37 ± 0.40>5
Ca2+ 28.4 ± 7.7718.444.312.95 ± 2.1235.7 ± 19.6917.684.030.42 ± 6.4936.7 ± 21.5716.091.629.85 ± 8.46200
Mg2+15.1 ± 4.719.722.97.86 ± 1.2914.4 ± 7.416.634.911.4 ± 2.4414.0 ± 5.517.424.87.6 ± 2.2050
Na+33.1 ± 15.7920.072.826.33 ± 4.3128.5 ± 11.1113.554.817.14 ± 3.6635.0 ± 14.6514.756.720.3 ± 5.88200
K+0.14 ± 0.090.070.290.15 ± 0.030.13 ± 0.090.050.380.15 ± 0.030.09 ± 0.060.010.180.09 ± 0.0312
NH4+0.02 ± 0.040.000.130.07 ± 0.010.02 ± 0.040.000.130.06 ± 0.010.02 ± 0.030.000.090.04 ± 0.011.50
S2−0.04 ± 0.020.020.070.03 ± 0.010.04 ± 0.030.020.130.04 ± 0.010.04 ± 0.020.020.070.03 ± 0.010.05
SO42−0.28 ± 0.200.010.590.34 ± 0.060.34 ± 0.250.011.020.39 ± 0.080.39 ± 0.290.021.180.40 ± 0.12250
Cl0.59 ± 0.400.011.670.66 ± 0.110.48 ± 0.340.011.400.53 ± 0.110.46 ± 0.380.001.540.53 ± 0.15250
CO32−0.13 ± 0.170.000.590.28 ± 0.050.08 ± 0.090.000.330.15 ± 0.030.13 ± 0.140.000.440.19 ± 0.06400
HCO30.11 ± 0.100.050.360.16 ± 0.030.11 ± 0.070.040.280.11 ± 0.020.13 ± 0.070.040.260.10 ± 0.03-
PO43−0.02 ± 0.020.000.060.03 ± 0.000.01 ± 0.010.000.060.02 ± 0.000.03 ± 0.020.000.060.03 ± 0.01-
COD1.94 ± 0.691.103.101.15 ± 0.192.31 ± 0.790.403.101.22 ± 0.261.81 ± 0.650.403.100.90 ± 0.26-
BOD0.24 ± 0.150.100.500.24 ± 0.040.21 ± 0.130.100.500.20 ± 0.040.22 ± 0.160.100.500.23 ± 0.07-
NO32.23 ± 2.000.307.503.33 ± 0.541.65 ± 1.500.105.602.33 ± 0.501.80 ± 1.690.105.602.34 ± 0.6850
NO20.14 ± 0.130.030.490.22 ± 0.040.08 ± 0.070.010.290.11 ± 0.020.13 ± 0.140.010.470.19 ± 0.053
F0.03 ± 0.020.010.060.03 ± 0.000.03 ± 0.020.000.090.03 ± 0.010.03 ± 0.030.010.080.04 ± 0.011.50
Mn0.19 ± 0.160.030.580.27 ± 0.040.13 ± 0.110.000.380.18 ± 0.040.19 ± 0.160.040.530.23 ± 0.070.40
Zn2.49 ± 1.420.905.002.36 ± 0.392.04 ± 1.340.205.002.08 ± 0.441.99 ± 1.300.904.101.80 ± 0.523.0
Pb0.02 ± 0.010.010.040.02 ± 0.000.01 ± 0.010.000.040.02 ± 0.000.09 ± 0.050.010.140.07 ± 0.020.01
Cd0.01 ± 0.010.000.020.01 ± 0.000.01 ± 0.010.000.010.01 ± 0.000.01 ± 0.010.000.020.01 ± 0.000.003
Cr0.02 ± 0.030.000.080.04 ± 0.010.02 ± 0.020.000.070.03 ± 0.010.02 ± 0.030.000.090.04 ± 0.010.05
Fe1.15 ± 0.830.332.901.39 ± 0.231.08 ± 0.740.093.001.15 ± 0.241.16 ± 0.900.143.101.24 ± 0.360.30
Ni0.02 ± 0.020.000.050.03 ± 0.010.01 ± 0.020.000.050.03 ± 0.010.01 ± 0.020.000.060.03 ± 0.010.07
TC0.65 ± 0.510.001.500.85 ± 0.140.60 ± 0.590.002.001.15 ± 0.250.65 ± 0.500.002.000.69 ± 0.20-
E. coli0.80 ± 0.900.003.001.50 ± 0.250.40 ± 0.550.002.000.84 ± 0.180.25 ± 0.450.001.500.63 ± 0.18-
WHO—World Health Organization, n = number of samples, Min = Minimum, Max = Maximum, SD = standard deviation, and SE = standard error, Temp = Temperature, TDS = total dissolve solid, TH = total hardness, EC = electric conductivity, DO = dissolved oxygen, Ca = calcium, Mg = magnesium, Na = sodium, K = potassium, NH4 = ammonium, S = sulfur, SO4 = sulfate, Cl = chlorine, HCO3 = bicarbonate, CO3 = carbonate, PO4 = phosphate, COD = chemical oxygen demand, BOD = biochemical oxygen demand, NO3 = nitrate, NO2 = nitrite, F = fluorine, Mn = manganese, Zn = zinc, Pb = lead, Cd = cadmium, Cr = chromium, Fe = iron, Ni = nickel, TC = total coliform, E. coli = Escherichia coli.
Table 3. Descriptive statistics of water quality parameters at residential, commercial, and industrial areas during rainy season.
Table 3. Descriptive statistics of water quality parameters at residential, commercial, and industrial areas during rainy season.
Water VariablesResidential (n = 15)Commercial (n = 25)Industrial (n = 10)WHO 2017
MeanMinMaxSD± SEMeanMinMaxSD± SEMeanMinMaxSD± SE
Temp (°C)28.0 ± 1.2025.030.02.00 ± 0.3328.4 ± 0.6626.030.01.03 ± 0.2228.9 ± 0.4928.030.00.68 ± 0.2-
pH7.7 ± 0.346.68.00.37 ± 0.067.1 ± 0.396.58.00.60 ± 0.137.3 ± 0.286.68.00.39 ± 0.116.5–8.5
TDS253.1 ± 108.69110.0440.0181.15 ± 29.67360.0 ± 169.98109.0750.0262.93 ± 56.15379.4 ± 178.86180.0761.0247.7 ± 71.86500
EC (µS/cm)412.2 ± 177.33179.0717.0295.52 ± 48.38585.9 ± 276.69177.01220.0428.4 ± 91.46617.4 ± 291.15293.01239.0403.2 ± 116.981250
Turbidity (NTU)2.9 ± 1.410.565.02.36 ± 0.392.1 ± 1.280.375.11.98 ± 0.423.6 ± 1.491.16.12.06 ± 0.605
TH142.8 ± 34.3791.3212.057.28 ± 9.38161.5 ± 54.4080.0273.084.09 ± 17.96169.0 ± 59.7182.3295.082.6 ± 23.97500
DO5.6 ± 1.113.97.91.85 ± 0.306.1 ± 1.222.19.51.88 ± 0.406.0 ± 1.034.08.01.42 ± 0.41>5
Ca2+42.1 ± 13.8521.670.023.09 ± 3.7850.9 ± 22.9024.2104.035.40 ± 7.5654.4 ± 28.6218.8112.039.6 ± 11.49200
Mg2+20.6 ± 5.5312.029.79.22 ± 1.5119.6 ± 9.517.042.314.68 ± 3.1418.9 ± 8.566.936.811.9 ± 3.4550
Na+38.1 ± 15.8320.078.726.40 ± 4.3245.0 ± 19.1417.588.129.60 ± 6.3247.3 ± 19.6920.583.527.2 ± 7.93200
K+0.23 ± 0.130.090.470.22 ± 0.040.24 ± 0.190.040.670.30 ± 0.060.13 ± 0.100.060.340.14 ± 0.0412
NH4+0.04 ± 0.060.000.190.10 ± 0.020.03 ± 0.050.000.190.08 ± 0.020.03 ± 0.040.000.110.05 ± 0.021.50
S2−0.07 ± 0.060.010.180.09 ± 0.020.07 ± 0.060.010.210.09 ± 0.020.09 ± 0.080.010.290.12 ± 0.030.05
SO42−0.42 ± 0.320.011.130.54 ± 0.090.53 ± 0.390.011.620.61 ± 0.130.58 ± 0.420.011.750.58 ± 0.17250
Cl0.85 ± 0.640.002.441.06 ± 0.170.76 ± 0.510.002.270.79 ± 0.170.65 ± 0.560.012.280.77 ± 0.23250
CO32−0.16 ± 0.190.000.650.32 ± 0.050.11 ± 0.120.000.430.19 ± 0.040.16 ± 0.180.000.610.25 ± 0.07400
HCO30.18 ± 0.140.050.500.23 ± 0.040.17 ± 0.120.050.530.18 ± 0.040.20 ± 0.100.050.440.14 ± 0.04-
PO43−0.04 ± 0.020.010.090.04 ± 0.010.04 ± 0.030.010.100.05 ± 0.010.05 ± 0.050.010.150.07 ± 0.02-
COD2