Abstract
Chemical elements, which are present in drinking water, could vary due to water sources, treatment processes or even the plumbing materials. Most of these elements do not represent a threat, while others, such as heavy metals, have been proven to cause harmful effects over human and aquatic wildlife. In this study, the quality of drinking water in three cities in Ecuador, Quito, Ibarra and Guayaquil was assessed through a multielement analysis and the heavy metal pollution index (HPI). A total of 102 drinking water samples and six natural water samples were collected and analyzed. Within the scope of analysis, results show that water quality complies with local and international guidelines. HPI did not show significant differences in the water that is supplied to the different neighborhoods of the three cities studied. However, actions should be taken to protect the sources of water, especially in Guayaquil, due to the presence of lead and chromium. For instance, lead was found in 2.8% of the samples in concentrations above World Health Organization (WHO) recommended values. Thus, we suggest to assessing the quality and age of the plumbing system within the whole country, in order to avoid drinking water contamination with heavy metals.
HIGHLIGHTS
Heavy metal pollution index (HPI) was obtained for Quito, Guayaquil and Ibarra. Results indicate a high HPI value for source water in Guayaquil related to the presence of lead and chromium, while Ibarra showed the lowest HPI for both source and drinking water.
Lead was also found in several samples in Quito, in concentrations above international guidelines (World Health Organization), possibly due to an aging plumbing system.
However, by considering only the analyzed elements, drinking water quality is similar in Quito and Guayaquil.
INTRODUCTION
Chemical compounds found in drinking water are classified according to their source in five main categories: naturally occurring, industrial sources, water treatment or materials used for plumbing and fittings, agricultural activities and pesticides. Elements commonly found in drinking water such as aluminum, sodium or potassium are present in chemical products used for the water treatment or in their water sources. Most of these products do not represent health risks and the guideline references are over 0.1 mg L−1. Other elements such as antimony, cadmium, copper, nickel, iron, zinc or lead are part of the materials used for plumbing and fittings, while arsenic, chromium, barium and manganese are found in water sources (WHO 2017). Table 1 summarizes commonly found elements in drinking water, their sources, guideline references and associated health risks according to the WHO (2017). Within this list, heavy metals are of special interest due to their health effects. This group of chemical elements owns unique properties, such as high density and the capacity to form water-soluble complexes. The high solubility of these complexes facilitates their transport and distribution to other strata in the food chain, such as soil, plants, groundwater and surface water, leading to their bioaccumulation in various organisms, especially the aquatic biome (Jaishankar et al. 2014; Pernía et al. 2018), and thus, contributing to their toxicity. The presence of heavy metals such as arsenic or chromium in drinking water is mainly due to its source (e.g. wells or groundwater). The stagnation of water in the distribution system and plumbing pipes might increase the concentrations of cadmium, copper, lead, iron or zinc (Chowdhury et al. 2016).
List of controlled parameters in drinking water
Parameter . | Main source . | Health risks . | WHO (2017) Guideline reference (mg L−1) . |
---|---|---|---|
Aluminuma | Water treatment | Available data inadequate | 0.1b |
Antimony | Plumbing and fittings | Antimony trioxide is possibly carcinogenic to humans | 0.02 |
Arsenicc | Naturally occurring | Carcinogenic | 0.01 |
Barium | Naturally occurring | Hypertension | 1.3 |
Boron | Natural and wastewater discharges | Testicular lesions have been observed in animals | 2.4 |
Cadmium | Plumbing and fittings, industrial | Possibly carcinogenic | 0.003 |
Chromiumd | Naturally occurring | Chromium (VI) human carcinogen | 0.05 |
Copper | Plumbing and fittings | Gastrointestinal effects | 2 |
Irona | Plumbing and fittings | Not of health concern at drinking water levels. Taste and appearance are affected under health limit | 1–3e |
Leadc | Plumbing and fittings | Possibly carcinogenic. Adverse neurotoxic effects | 0.01 |
Manganesea | Naturally occurring | Alleged neurological effects | 0.4e |
Mercury (Inorganic) | Industrial | Hemorrhagic gastritis and colitis, kidney damage. Possibly carcinogenic | 0.006 |
Molybdenuma | Naturally occurring, agricultural, industrial | No data are available | 0.07e |
Nickel | Naturally occurring/plumbing and fittings | Allergic contact dermatitis. Possibly carcinogenic | 0.07 |
Potassiuma | Naturally occurring/water treatment | Not of health concern at drinking water levels. Taste is affected under health limit | 0.2e |
Seleniumd | Naturally occurring | Toxic effects of long-term selenium exposure are manifest in nails, hair and liver | 0.04 |
Silvera | Naturally occurring/water treatment | Available data inadequate | 0.1e |
Sodiuma | Naturally occurring/water treatment | Not of health concern at drinking water levels | 200e |
Zinca | Plumbing and fittings | Not of health concern at drinking water levels | 3e |
Parameter . | Main source . | Health risks . | WHO (2017) Guideline reference (mg L−1) . |
---|---|---|---|
Aluminuma | Water treatment | Available data inadequate | 0.1b |
Antimony | Plumbing and fittings | Antimony trioxide is possibly carcinogenic to humans | 0.02 |
Arsenicc | Naturally occurring | Carcinogenic | 0.01 |
Barium | Naturally occurring | Hypertension | 1.3 |
Boron | Natural and wastewater discharges | Testicular lesions have been observed in animals | 2.4 |
Cadmium | Plumbing and fittings, industrial | Possibly carcinogenic | 0.003 |
Chromiumd | Naturally occurring | Chromium (VI) human carcinogen | 0.05 |
Copper | Plumbing and fittings | Gastrointestinal effects | 2 |
Irona | Plumbing and fittings | Not of health concern at drinking water levels. Taste and appearance are affected under health limit | 1–3e |
Leadc | Plumbing and fittings | Possibly carcinogenic. Adverse neurotoxic effects | 0.01 |
Manganesea | Naturally occurring | Alleged neurological effects | 0.4e |
Mercury (Inorganic) | Industrial | Hemorrhagic gastritis and colitis, kidney damage. Possibly carcinogenic | 0.006 |
Molybdenuma | Naturally occurring, agricultural, industrial | No data are available | 0.07e |
Nickel | Naturally occurring/plumbing and fittings | Allergic contact dermatitis. Possibly carcinogenic | 0.07 |
Potassiuma | Naturally occurring/water treatment | Not of health concern at drinking water levels. Taste is affected under health limit | 0.2e |
Seleniumd | Naturally occurring | Toxic effects of long-term selenium exposure are manifest in nails, hair and liver | 0.04 |
Silvera | Naturally occurring/water treatment | Available data inadequate | 0.1e |
Sodiuma | Naturally occurring/water treatment | Not of health concern at drinking water levels | 200e |
Zinca | Plumbing and fittings | Not of health concern at drinking water levels | 3e |
aGuideline values have not been established due to not presenting health risks at drinking water levels or due to tasting and/or appearance effects are below health limit.
b0.2 mg L−1 for small facilities.
cThe guideline value is provisional due to the achievable levels by analysis techniques or treatment methods.
dThe guideline value is provisional due to inadequate health database.
eReferential values and guideline values have not been established (WHO 2017).
According to a study carried out by the United Nations Children's Fund (UNICEF) and the Ecuadorian National Institute of Statistics (INEC in Spanish), 70% of the population in Ecuador has access to safe water. These figures are based on indicators such as water quality and quantity, closeness to water source and distribution systems (Molina et al. 2018). The Ecuadorian population is divided into 62% urban and 38% rural. Only 74% of the rural population has access to a household installation compared with 96% of the urban population (INEC 2016). The main water sources in Ecuador are rivers from mountainous origin, which flow out into the Pacific Ocean in the west and into the Amazon River in the east. According to the Economic Commission for Latin America and the Caribbean (CEPAL 2012), potabilization processes in Ecuador are conventional, usually following a route of aeration, coagulation, flocculation, sedimentation, filtration and disinfection with chlorine; these processes are adjusted according to conditions of raw water or seasonal changes.
Studies performed in Ecuador have determined the presence of several heavy metals in matrices such as surface waters, sediments, fish, soils, crops, particulate matter (PM10), among others (Correa-Cruz et al. 2015; Barraza et al. 2018; Pernía et al. 2018). In some cases, values exceed the corresponding regulations, posing a high risk to the ecosystem and the population (Araujo & Cedeño-Macías 2016; Barraza et al. 2018). Up to now, no large-scale epidemiological studies have been conducted in Ecuador; however, some studies report a health risk for the population due to the drinking water quality (Maurice et al. 2019). For instance, Hernández-Bonilla et al. (2016) indicate a negative effect on the IQ of schoolchildren related to the presence of manganese, lead and mercury coming from mining in the Paquisha region. In another study, biological markers such as methyl mercury and lead were measured in hair and blood, respectively, showing results significantly higher than recommended values (Correa-Cruz et al. 2015). Water pollution cases have been reported in the main cities of Ecuador. In the Metropolitan District of Quito, at Tumbaco neighborhood, arsenic levels ten times higher than WHO guidelines for drinking water were detected in 2004 (García 2012). The existence of arsenic in well water from Papallacta and Pifo has also been determined (Alulema 2017). In Guayaquil, water pollution occurs in the Daule River due to untreated domestic and industrial wastewater discharges, poor management of solid waste, as well as fuel spills (Da Ros 1995). Meanwhile, in the city of Ibarra, there are no studies nor official publications of heavy metals in the water supplied for human consumption.
The aims of this study were to determine the quality of drinking water in three cities of Ecuador based on the multi-element content and the Heavy Metal Pollution Index (HPI). Selected cities have differences in their water catchment systems and the extent of potabilization processes, which provides a full picture of the variations found in the urban areas of Ecuador. The two most populated cities (Quito and Guayaquil) were included. We hypothesize that, among cities, variations in the chemical composition of the source water, as well as their treatment and distribution systems, will lead to differences in the quality of the water supplied to the urban population. In addition, through the HPI values, a preliminary assessment of the presence of heavy metals in water is intended, as these elements have a high associated risk to individual and public health.
MATERIALS AND METHODS
Study area
This study was carried out in three cities of Ecuador: Ibarra, Quito and Guayaquil (Figure 1). Ibarra is the capital city of the Imbabura province in northern Ecuador; located at an elevation of 2,225 m on the banks of the Tahuando River. Ibarra is a medium-sized city with around 130,000 inhabitants over a 41 km2 area. This city has two main water catchments zones. According to the Municipal Public Water and Sanitation Company of Ibarra (EMAPA-I 2005), the Palestina river watershed is the source for the sectors of Florida, Ejido de Caranqui, Pugacho Alto and Bajo, while groundwater from the Guaraczapas river supplies water to the rest of the municipality, including rural parishes such as La Esperanza and San Antonio. Quito is the capital city of Ecuador and the Pichincha province, located at an elevation of 2,850 m on the hillsides of the Pichincha volcano in the western side on the Andes. It is the second-largest city in Ecuador with about 2,000,000 inhabitants over a 372 km2 area. The sources of water for the capital are nearby lakes, wells and springs. The 98% of this water goes through potabilization treatment before distribution, with a total production of ≈8 m3/s (Metropolitan Public Water & Sanitation Company of Quito (EPMAPS-Q) 2011). Guayaquil is the capital of the Guayas province, located at the sea level on the Pacific coast at the mouth of the Guayas River. It is the most populated city in Ecuador with about 4,000,000 inhabitants over a 347 km2 area. In Guayaquil, around 90% of the drinking water is distributed by four water supply systems, which are linked to the pumping station of ‘La Toma’ on the Daule River. The water is treated at ‘La Toma’ and distributed to the city (Interagua 2011).
Sampling sites in Ibarra, Quito and Guayaquil. Location of sectors and/or parishes considered for sampling are represented by grayscale; black bullets indicate the sampling points in each sector/parish. In addition, the location of each city within the Ecuadorian territory is shown.
Sampling sites in Ibarra, Quito and Guayaquil. Location of sectors and/or parishes considered for sampling are represented by grayscale; black bullets indicate the sampling points in each sector/parish. In addition, the location of each city within the Ecuadorian territory is shown.
Potabilization treatment
Mostly, a conventional process is carried out in Ibarra, Quito and Guayaquil. Briefly, it consists of removal of solids, filtration and disinfection of the water. A flocculation–coagulation process is used to remove the solids; the most widely used coagulant is aluminum sulfate, mainly due to its low price. Mixing and flocculation are carried out through hydraulic methods. Sedimentation occurs in decanters, generating the floc precipitation, and thus forming sludge. Subsequently, water is filtrated to remove suspended solids on a porous medium, finally, a chlorination–disinfection process is performed to eliminate pathogenic microorganisms.
Sampling sites
Sampling sites for the three cities were established according to their distribution systems, dividing the city into parishes and/or neighborhoods, and establishing three random sampling points within each zone (Figure 1). In the city of Ibarra, 15 sampling points were established by considering the five urban parishes of the municipality: Alpachaca (n = 3), El Sagrario (n = 3), San Francisco (n = 3), Caranqui (n = 3) and Priorato (n = 3) (Figure 1). In addition, three replicate samples were taken at the intake of the water purification plant (WPP) of the Guaraczapas sector. In Quito, 45 sampling points were established in 15 neighborhoods: six in Bellavista (n = 18), three in Puengasí (n = 9), three in El Placer (n = 9) and three in El Troje (n = 9) (Figure 1). Collection of samples at the water sources in Quito was not possible due to restricted access. In the city of Guayaquil, 42 sampling points were established as follows: five neighborhoods in the southern zone (n = 12), four neighborhoods in the central zone (n = 15) and three neighborhoods in the northern zone (n = 15) (Figure 1). In Alborada and Urdesa, neighborhoods of the Northern zone, six sampling points were taken due to their extension. Three additional sampling points were taken at the intake zone for La Toma Complex – drinking water treatment plant in the Daule River. A total of 108 samples was analyzed.
Fieldwork
Sampling protocol was performed as described in Standard Methods 1060 (2017). Samples were taken directly from the distribution system in faucets located in home kitchens or bathrooms. Plastic containers of 100 mL were used for collection and samples were preserved in situ with nitric acid at pH less than 2, stored in containers with ice and transported to the laboratory. Samples were stored at 4 °C until their respective analysis. Additionally, daily field blanks and travel blanks were taken during the sampling as quality controls. Blanks were used as samples and analyzed to discard any contamination that might occur during the sampling, handling or transportation of the samples to the laboratory.
At each sampling point, the following in situ parameters were determined with previously calibrated field instruments: pH and temperature (Martini MI805), conductivity (Metrohm Conductometer 912), turbidity and free chlorine (HACH DR890). All material was rinsed four times with deionized water (18.2 MΩ cm−1) to avoid cross-contamination between sampling points. Exclusive containers were used to measure the parameters in situ to avoid contamination of the containers used for measuring metals and other parameters.
Chemical analyses
Major elements and trace metals were analyzed in acidified water samples by means of an Atomic Emission Spectrometer with Inductive Plasma Coupling (ICP-OES Thermo Scientific iCAP 7400) at the Laboratory of Environmental Engineering – Universidad San Francisco de Quito (LIA – USFQ), Ecuador. Calibration curves were prepared from an ICP Merck-Millipore multi-element standard solution VIII – mixture of several analytes at a concentration of 100 mg L−1 (grade Certipur for ICP, Merck-Millipore). Limits of detection (LOD) and quantification (LOQ) were calculated by analyzing at least 12 independent replicates of blank samples and multiplying the standard deviation by three and by ten to obtain the LOD and LOQ, respectively.
Quality control for water analysis was performed using a certified reference material (CRM 1640a NIST, Gaithersburg, MD) every 15 samples. Recovery percentages were calculated to determine matrix effects and method accuracy; the values obtained were between 86 and 110%. Concentrations of every metal were corrected according to the percentage of recovery for each analyte. The elements analyzed in this study were aluminum (Al), arsenic (As), barium (Ba), bismuth (Bi), calcium (Ca), cadmium (Cd), cobalt (Co), copper (Cu), chromium (Cr), iron (Fe), potassium (K), magnesium (Mg), manganese (Mn), sodium (Na), nickel (Ni), lead (Pb), strontium (Sr) and zinc (Zn). Additionally, samples with high metal content were sent to an external accredited laboratory to double-check the obtained values.
Heavy metal pollution index
Equation (2) results from modifying Equations (2) and (5) from Chaturvedi et al. (2019) by multiplying the partial heavy pollution index by 100 (Vetrimurugan et al. 2016) and considering the maximum desirable concentration of every metal as 0. Heavy metals were selected as a function of the potential effects on human health and the maximum permissible concentrations (Tables 1 and 2).
Selected heavy metals, reference values and calculated weights for the calculation of the HPI
Metal . | Reference value (mg L−1) . | Weight . |
---|---|---|
Barium | 0.7 | 0.010 |
Chromium | 0.05 | 0.282 |
Copper | 2 | 0.004 |
Lead | 0.01 | 0.705 |
Metal . | Reference value (mg L−1) . | Weight . |
---|---|---|
Barium | 0.7 | 0.010 |
Chromium | 0.05 | 0.282 |
Copper | 2 | 0.004 |
Lead | 0.01 | 0.705 |
All standard reference values have been taken from Ecuadorian guidelines (INEN 2014).
Statistical analyses
Values below LOD were substituted by half the value of the detection level for plotting the figures and for statistical analyses. Differences in the chemical composition of the water sources, as well as raw and treated water between Ibarra and Guayaquil, were compared with the Wilcoxon test. Differences in the chemical composition of treated (drinking) water among the three cities were tested with the Kruskal–Wallis test and the Wilcoxon test as a non-parametric multiple comparison method. The probability of a Wilcoxon test was adjusted to consider multiple comparisons (Benjamini & Yekutieli 2001). Differences in HPI among source water and the water supplied to the different neighborhoods were also tested with the Kruskal–Wallis and Wilcoxon tests.
Pairwise Pearson correlations among all the measured variables were performed and a Bonferroni correction was applied to the significance level. Data were centered (mean subtracted from data) and scaled (data divided by standard deviation). Principal component analysis (PCA) was performed to search for variable correlations and groupings of samples. Significance of variable correlation with the obtained principal components was tested using the broken-stick method (Peres-Neto et al. 2003). All statistical analyses were executed in R (R Core Team 2019).
RESULTS
Water sources
In terms of the chemistry of the water sources, no significant differences were observed between Ibarra and Guayaquil, probably because the low sample size resulted in a low sensitivity of the Wilcoxon test. Higher concentrations of Al, Ba, Cr, Fe, Mn, Pb and Zn were observed in Guayaquil and Sr was the only heavy metal with higher concentration in Ibarra (Figure 2). On the other hand, concentrations of Ca, Mg and Na were higher in Ibarra, while the concentrations of K were similar in both cities (Figure 2).
Element concentrations (mean ± standard deviation) in water sources of Ibarra (IBA) and Guayaquil (GUA). Results of the Wilcoxon test (*p < 0.05; **p < 0.01; ***p < 0.001) for differences between the source and treated water of both cities are also shown.
Element concentrations (mean ± standard deviation) in water sources of Ibarra (IBA) and Guayaquil (GUA). Results of the Wilcoxon test (*p < 0.05; **p < 0.01; ***p < 0.001) for differences between the source and treated water of both cities are also shown.
Comparison of source and treated water
In Ibarra, concentrations of Ba, Cu and Zn in the treated water were significantly higher than in the source water (Figure 2). There were also significantly higher Al concentrations in the treated water, but the difference was negligible. Mg and Mn concentrations in the source were higher than in the treated water, but differences were also small. Finally, no significant differences between the source water and the treated water were observed for Ca, Cr, Fe, K, Na, Pb and Sr. In Guayaquil, concentrations of Al, Cr, Fe, Mn, Pb and Sr in the source water were significantly higher than in the treated water (Figure 2). Conversely, concentrations of Ca, Cu and Na significantly increased in the treated water. Other elements including Ba, K, Mg and Zn showed no significant differences between source and treated water.
Drinking water comparison
The results of treated water were compared between the three cities, and significant differences were observed in pH, conductivity, turbidity and free chlorine (Figure 3). The pH and turbidity were higher in Guayaquil, while conductivity and free chlorine were lower compared to the other cities. Ibarra was the city with higher conductivity values and Quito showed significantly higher concentrations of free chlorine compared to Guayaquil and Ibarra.
Physical and chemical characteristics of treated water in Ibarra, Quito and Guayaquil. Results of the Kruskal–Wallis test (*p < 0.05; **p < 0.01; ***p < 0.001) and the Wilcoxon test for differences among the cities are also shown. There are no significant differences between levels with the same letter.
Physical and chemical characteristics of treated water in Ibarra, Quito and Guayaquil. Results of the Kruskal–Wallis test (*p < 0.05; **p < 0.01; ***p < 0.001) and the Wilcoxon test for differences among the cities are also shown. There are no significant differences between levels with the same letter.
All non-heavy elements showed significant differences between the three cities, except for K (Figure 3). Other elements could be grouped based on their concentration pattern by city. Al and Fe followed the same pattern: the city with the highest concentrations was Quito, followed by Guayaquil and Ibarra. Ba, Ca, Mg, Na and Sr showed similar results, being Ibarra the city with the highest concentrations. Finally, Mn was significantly higher in Quito, compared to Ibarra and Guayaquil. The heavy metals analyzed (As, Cd, Cr and Pb) showed concentrations below the limit of detection (As LOD: 9.67 μg L−1, Cd LOD: 0.58 μg L−1, Cr LOD: 1.24 μg L−1, Pb LOD: 5.96 μg L−1) except for three samples in Quito where concentrations above World Health Organization recommended values were found. Cu and Zn were present in all cases; however, Quito had the highest variation of metal concentration. Cu concentration in Quito showed significant differences compared to Guayaquil but not with Ibarra. In addition, Zn concentration showed statistically higher values in Ibarra compared to Quito and Guayaquil (Figure 3).
HPI values
HPI showed contrasting values among the three cities (Figure 4). In Ibarra, there was a significant difference between the water at the intake of the distribution system and the treated water, but it was negligible (21.4 at the intake and 21.8 in the supply water). In Quito, there were no significant differences in the index values among the water supplied to the different sectors of the city. However, three samples showed values in the range 133–490, which lead to high variability of the HPI in the sectors of Puengasi, El Troje and Noroccidente. In Guayaquil, HPI at the intake on the Daule river was significantly higher than in the treated water. No significant differences in the HPI in the water that is supplied to the different sectors of the city were observed.
HPI of water collected at the intake of the water distribution system and at different sectors in Ibarra, Quito and Guayaquil. Results of the Kruskal–Wallis test (*p < 0.05; **p < 0.01; ***p < 0.001) and the Wilcoxon test for differences among sectors within each city are shown. There are no significant differences between levels with the same letter.
HPI of water collected at the intake of the water distribution system and at different sectors in Ibarra, Quito and Guayaquil. Results of the Kruskal–Wallis test (*p < 0.05; **p < 0.01; ***p < 0.001) and the Wilcoxon test for differences among sectors within each city are shown. There are no significant differences between levels with the same letter.
Correlation and PCA analysis
The measured variables were grouped into three groups as a function of their pairwise correlations (Supplementary Material, Figure S1 and Table S1). The first group comprised turbidity, Al, Fe, Cr, Mn and Pb and showed positive correlations with factor values between 0.606 and 0.985 (Mn and Pb: n = 108, r = 0.660, p < 0.001; Fe and Cr: n = 108, r = 0.985, p < 0.001). The second group comprised conductivity, Ba, Ca, K, Mg, Na, Sr and Zn and showed positive correlations between 0.447 and 0.972 (Mn and Pb: n = 108, r = 0.660, p < 0.001; Fe and Cr: n = 108, r = 0.985, p < 0.001). Finally, pH and free chlorine concentration were negatively correlated (n = 108, r = − 0.648, p < 0.001).
The first three components of the PCA, shown in Figure 5, explained 79% of the variability in the samples, as detailed in Table 3. The first principal component explained 40% of the variability and segregated the samples of Ibarra from the other two cities. It also segregated the samples of the water potabilization plant intake at the Daule river from the samples collected in Guayaquil. The position of the Ibarra samples along this axis was explained by higher conductivity values and higher concentrations of Ca (Table 3); a single sample from the Bellavista neighborhood with high Ca concentration also appeared segregated from other samples collected at Quito. The second principal component explained 27% of the variability and segregated the samples of Bellavista, El Troje and the Daule river from the other samples; this was explained by higher concentrations of Al and Cr (Table 3). The third principal component explained 12% of the variability and segregated samples of Ibarra and Guayaquil from the samples collected at Quito. The position of the Quito samples along the third axis was explained by lower pH and higher chlorine concentrations (Table 3); the third axis also segregated samples collected at the outlet of the WPP from other samples collected in Ibarra. Most of the Guayaquil and Quito samples were overlapped along the first and second axes of the PCA space (Figure 5).
Results of PCA on the chemical analysis of water samples from Ibarra, Guayaquil and Quito
. | PC1 . | PC2 . | PC3 . |
---|---|---|---|
Variance (%) | 39.5 | 27.4 | 12.1 |
Cumulative variance (%) | 39.5 | 66.9 | 79.0 |
pH | – | – | −0.544 |
Conductivity | −0.339 | – | – |
Turbidity | – | – | – |
Free chlorine | – | – | 0.581 |
Al | – | −0.350 | – |
Ba | – | – | – |
Ca | −0.335 | – | – |
Fe | – | – | – |
K | – | – | – |
Mg | – | – | – |
Mn | – | – | – |
Na | – | – | – |
Sr | – | – | – |
Cu | – | – | – |
Cr | – | −0.378 | – |
Pb | – | – | – |
Zn | – | – | – |
. | PC1 . | PC2 . | PC3 . |
---|---|---|---|
Variance (%) | 39.5 | 27.4 | 12.1 |
Cumulative variance (%) | 39.5 | 66.9 | 79.0 |
pH | – | – | −0.544 |
Conductivity | −0.339 | – | – |
Turbidity | – | – | – |
Free chlorine | – | – | 0.581 |
Al | – | −0.350 | – |
Ba | – | – | – |
Ca | −0.335 | – | – |
Fe | – | – | – |
K | – | – | – |
Mg | – | – | – |
Mn | – | – | – |
Na | – | – | – |
Sr | – | – | – |
Cu | – | – | – |
Cr | – | −0.378 | – |
Pb | – | – | – |
Zn | – | – | – |
Variance and cumulative variance are explained by the selected components; significant variable loadings are shown.
Results of the PCA on the chemical analysis of supply waters of Ibarra, Guayaquil and Quito. Most of the Guayaquil and Quito samples are overlapped.
Results of the PCA on the chemical analysis of supply waters of Ibarra, Guayaquil and Quito. Most of the Guayaquil and Quito samples are overlapped.
DISCUSSION
Overall, the results match the occurrence of the main kind of elements present in drinking water in other developing countries; elements such as Pb, Ni, Zn, Ba, Fe, Cr and Cu were found in this study and are also reported by Chowdhury et al. (2016). Despite finding some heavy metals in certain source water samples, it seems that potabilization treatments applied are effective in decreasing their concentration and provide a product that meets all sanitary national and international regulations. However, distribution networks could be identified as the source of some chemical elements such as Cu or Pb.
The correlations among variables and the PCA highlighted the differences in the chemical composition of supplied water among the three cities. As shown in Figure 3, treated water in Ibarra has higher conductivity than in Quito and Guayaquil and high concentrations of Ba, Ca, Mg, Na, Sr and Zn, while drinking water in Quito and Guayaquil has higher Al and Fe concentrations. Differences in the chemical composition of treated water result from the use of different water sources, groundwater in Ibarra and surface water in Quito and Guayaquil. This analysis also highlighted differences in pH and chlorination among the cities, specifically, that water distributed in Quito has lower pH and higher free chlorine concentrations. Also, most water samples from Quito and Guayaquil were overlapped on the PCA (Figure 5), which indicates that drinking water quality is similar in both cities, considering the parameters analyzed in this study.
The source water in Ibarra is less polluted than in Guayaquil, showing lower concentrations of metals such as Cr, Cd or Pb (Figure 2). This difference could be explained by the fact that the Daule river receives industrial and household wastewater discharges, usually without prior treatment. Ibarra has higher concentrations of metals commonly found in groundwater (Ca, Mg, Na and Sr) because these waters are exposed to different chemical processes during rock–water interaction, such as dissolution, precipitation, ion exchange and reduction (Elango & Kannan 2007).
After treatment and distribution of the water, physicochemical parameters like conductivity and turbidity resulted in normal values for drinking water within the three studied cities. Nevertheless, this was not the case for some results of pH and free chlorine. The WHO recommends 6.5–8 as the optimal range for pH; although these values are not based on health risks, the pH of the water entering into the distribution system must be controlled to minimize the corrosion of water pipes both in municipal and household systems (Li et al. 2016). Very low pH values can cause corrosion in the pipes, while very high values can affect the effectiveness of the disinfection. The pH values found in 50% of sampling points in Quito were below 6.5, which could be attributed to plant treatment operational conditions to ensure the concentration of free chlorine. High concentrations of chlorine decrease the pH value, according to the results the presence of free chlorine was higher in this city compared to the other two; thus, the results are consistent. The value of free chlorine in the samples from Quito and Ibarra was within the range established in the national regulation: 0.3–1.5 mg L−1 (INEN 1108 2014). On the other hand, in many cases, free chlorine measurements in the city of Guayaquil were below 0.3 mg L−1, which could pose a risk to the population since lower levels allow the proliferation of microorganisms, and consequently, an increased occurrence of gastrointestinal diseases. These low values could be explained by the distance from the treatment plant to some city sectors, as the main problem of all water supply systems is the loss of water stability during the distribution to final users. In addition, as chlorine reacts with organic and inorganic compounds in water, its concentration over time decreases, a phenomenon called ‘chlorine decay’ (Boccelli et al. 2003). This decay could also explain the low levels found in Guayaquil, as the source water has a high concentration of organic matter. Murphy et al. (2015) indicate that, even in developed countries, waterborne diseases can be originated from contamination during drinking water distribution. A poor maintenance of the distribution system can act as a vehicle of transmission for pathogens and may even significantly contribute to gastrointestinal disease in the community (Lee & Schwab 2005). Special care should be taken on this issue to ensure proper water disinfection for the city of Guayaquil.
Metals content in water depends on several parameters including geological origin and treatments performed as part of the potabilization process. Thus, most of the elements showed significant differences in their concentrations among the three studied cities. It is important to notice that the drinking water of Quito, Ibarra and Guayaquil is soft; Ca and Mg are present in small amounts, which might contribute in a positive way to the daily intake of these minerals, as they have benefits in the reduction of cardiovascular mortality rates (WHO 2009). Sr, Ba and Zn also contribute with hardness, but the found values are low and do not have effects over human health. Regarding Fe and Mn, it is known that above certain concentrations they affect the taste of the water (WHO 2017); however, the amounts observed in this study are low enough and far from becoming a concern.
Other metals occurrence in water supply is due to plumbing and fittings materials used in the distribution network from treatment plants to final users (WHO 2017). Although the concentrations of some heavy metals in treated water are extremely low or even undetectable, they can accumulate and concentrate in loose deposits over time. When the water quality conditions change, the heavy metals in the loose deposits could be released back into the bulk water at a significant concentration and place consumer health at a high risk (Li et al. 2016). In Ibarra, a contamination with Cu and Zn after the treatment plant is evident. According to the Ecuadorian Water Secretariat (Secretaría del Agua 2015), the copper was a material commonly employed in water pipelines, and the treated water consumption meters have a copper coating; therefore, it is very likely that small traces of this element infiltrate the stream of water. An increased concentration of this element is also found in the water supply system of Guayaquil, possibly due to the same reasons. Although copper is a necessary component for normal metabolism in humans, acute poisoning from ingestion of excessive copper can cause temporary gastrointestinal distress (nausea, vomiting and abdominal pain) (Jaishankar et al. 2014). However, the obtained concentrations of Cu and Zn in Ibarra do not represent a risk for health, and they are under the permitted guidelines established by INEN (2014) and WHO (2017). Furthermore, the WHO (2017) does not establish a reference value for Zn because it does not pose a health problem at the levels found in drinking water.
Without a doubt, the state, maintenance and type of the pipe system have an important influence in the quality of water. In this study, some findings suggest a possible drinking water contamination with lead or copper due to the pipes and fittings. The water distribution network of Ibarra was constructed after 1970, so the presence of lead pipes and fittings is unlikely; the main pipe system is made of polyvinyl chloride (PVC), but there are still some asbestos concrete pipes which are susceptible to leaks, which could potentially lead to contaminations (EMAPA-I 2005). Construction of the water distribution network of Guayaquil started in 1946 and the water treatment plant of La Toma, which is still providing water to the city, was inaugurated in 1950. However, as most of the pipe system is made of PVC (Interagua 2011), lead concentrations in drinking water were below detection levels in all samples. In the north and south of Quito, the networks were built in the 1960s, when the first municipal drinking water company was established; although most of the pipeline is made of PVC (EPMAPS-Q 2020), it is possible that old lead plumbing systems are still in use in some neighborhoods; hence, the heavy metal presence found in this study.
Distribution system deficiencies, such as inadequate disinfection residual, low pressure, intermittent service, leaks and corrosion, often interact with each other, decreasing the quantity and quality of water arriving at the household (Lee & Schwab 2005). A decrease in water quality can also be perceived with HPI values (Figure 4). These indicate that Ibarra is the city with the best water quality, probably because the main water source is underground and there is no presence of heavy metals before nor after water distribution into the city. In Quito, the high variability in El Troje is due to high concentrations of Pb.
Increased HPI variation was also observed in Puengasí and Noroccidente, which may indicate pipe corrosion in these three neighborhoods located in the south of the city. Considering that 6.6% of the samples of Quito presented lead levels between two and six times higher than the guideline, it is highly recommended to perform a more extensive research to evaluate the state of the plumbing system, particularly in the southern sector, because lead is a neurotoxic heavy metal, and it is necessary to avoid any risk for the exposed population. The WHO (2017) ensures that exposure to lead is associated with a wide range of effects, including various neurodevelopmental effects, mortality (mainly due to cardiovascular diseases), impaired renal function, hypertension, impaired fertility and adverse pregnancy outcomes. According to the International Agency for Research on Cancer (IARC 2019), stomach cancer is the first cause of cancer death in Ecuadorians with 2,085 (14.3%) new deaths recorded in 2018. Some studies looking at blood lead levels in the population have found an increased risk of stomach cancer associated with higher lead exposure. Although Pb does not usually cause acute poisoning, the link between Pb exposure for a long period and cancer is clearly a concern (Jaishankar et al. 2014). Lead exposure cannot be related only to drinking water consumption since there are other sources of contact such as the workplace or through air or food (Jaishankar et al. 2014).
In Guayaquil, the HPI confirms the results of the PCA; the concentrations of metals in the Daule river are much higher than those found in the city. Although distributed water is within the parameters established by the WHO, it is necessary to assess the effectiveness of current environmental laws to ensure clean water access for the population of Guayaquil city, as well as the sustainability of the catchment zones.
CONCLUSIONS
Results obtained from the physicochemical parameters and multielement analysis of drinking water in the cities of Quito, Ibarra and Guayaquil showed that, overall, water distributed to consumers complies with the parameters established in both international and national regulations. However, based on our observations, there are a few actions that should be taken in each city. As high values of lead were found in three neighborhoods in Quito, it is necessary to conduct more analyses to assess their origin and health risk to the population since lead is a highly toxic metal and has been linked to various types of cancer. In addition, measures should be taken to avoid threats to water catchment zones in Guayaquil, where a high HPI was found, in order to reduce future operational costs for the treatment facilities. Finally, based on the concentrations of copper and zinc in Ibarra, it would be advisable to renew the pipes from the distribution network; even though they do not represent a risk for health, these metals are indicators of corrosion.
FINANCIAL SUPPORT
This study was conducted as part of the Research Project O13025, sponsored by Pontificia Universidad Católica del Ecuador (PUCE). Financing was also provided by the Universidad San Francisco de Quito (USFQ) through the program POLI-grant 2017. Neither PUCE nor USFQ, as institutions, was involved in data collection and analysis, nor publication preparation, which is the sole responsibility of the authors.
DATA AVAILABILITY STATEMENT
All relevant data are included in the paper or its Supplementary Information.