Assessment of drinking water suitability in low income rural areas: a case study in Sixaola, Costa Rica

Vegas-Las Palmas is a rural settlement located in the southern Caribbean region of Costa Rica on the border with Panama. Its population does not have access to potable water, and inhabitants depend on water from wells at the water table level to meet their needs. These wells lack basic infrastructure to protect this water from contamination. In this study, water quality was evaluated at 12 wells from 2014 to 2016 (n1⁄4 72). The results revealed high concentrations of faecal coliforms and Escherichia coli with maximum values of 4.6 × 10 MPN/100 mL and 1.1 × 10 MPN/100 mL, respectively. In addition, maximum values of pH, conductivity, turbidity, Ca, Mg, K, Fe, Mn, Cd and Pb were found to be outside the standard limits (nationally and internationally) for potable water. Possible sources of water contamination are associated with the geomorphological characteristics of the area, as well as with hydrometeorological and anthropogenic factors such as the lack of sewerage, the presence of latrines, animals near the wells and the use of agrochemicals. The water quality was heterogeneous among wells, and all of them were found to have conditions that caused water to be unfit for human


INTRODUCTION
Access to safe and good quality drinking water is a human right (Montreal International Forum ).Everyone has the right to water in sufficient quantity to meet their needs because it is essential for human development, health, and well-being (Varol & Dravaz ).In 2015, the Millennium Development Goals target of 88% of the population having access to improved sources of drinking water was achieved (WHO ).However, approximately 663 million people around the world still lack basic access to safe drinking water (Fanucchi ).Water supply per person should be continuous and sufficient for personal and domestic use but must also be free of microorganisms, chemical substances and radiological hazards which pose a threat to human health (UN ).The lack of drinkable water, inadequate management of water resources, presence of pathogenic organisms, and lack of sanitation and hygiene eventually increase the presence of diseases that can affect human health (Sorline et al. ; Varol & Dravaz ).
Contamination of drinking water is caused by natural or anthropogenic factors, which can be present between the source and the distribution point, therefore increasing the possibility of transmitting diseases (Valiente & Mora ).
Trevett et al. () mention a wide number of critical points in the capture, transport, storage, distribution, and use of water during which contamination could be introducedfor example, hands, containers, ladles, filter cloths, dust, insects and animals are potential sources of contamination that could lead to the presence of bacteria or other pathogenic organisms in the water.In order to reduce the impacts of these sources of pollution, better management of the water supply is necessary, incorporating good practices from the source of supply to the consumer (Sorline et al. ).
In addition, extreme hydrometeorological changes caused by climate change may affect water quality, especially surface water ( Jovanelly et al. ), by altering physical and chemical variables including pH, alkalinity and temperature (Agudelo ; Moreira & Bondelind ).
Developing countries have made significant progress in access to and coverage of potable water.Particularly in Costa Rica, the percentage of the population with access to water in their homes was 99.5% in 2014, with a significant increase during the previous years; but only 93.4% of the population has access to safe drinking water (State of the Nation Program , ).Thus, there are still marginal communities that do not have access to potable water, many of which are in areas of high vulnerability to natural disasters and with limited access to public services (Bower ).This information contrasts with Article 50 of the Constitution (), which states that access to a sanitary water supply is a fundamental right derived from the right to life and health, and should be seen as a public service that cannot be denied, through either inaction by the government or the actions of the provider.

One of this communities enduring difficult conditions is
Vegas-Las Palmas parcels in Sixaola.There, the inhabitants obtain water of uncertain quality and quantity from different sources.These sources include wells in poor condition, river water and rainwater; none of this water undergoes treatment that guarantees its fitness for human consumption.The present study carried out a physical, chemical, and microbiological assessment of well water to define a baseline that would identify the possible risks which inhabitants are exposed to.This information is essential due to the prevailing conditions affecting water supply for human consumption in the community and for implementing mitigation measures to improve the health of inhabitants.

Study area
Vegas-Las Palmas parcels are located in the Sixaola district of the canton of Talamanca, in the province of Limón, Costa Rica, between coordinates 9 33 0 0″ N and 82 36 0 0″ W and 9 30 0 0″ N and 82 38 0 0″ W (Figure 1).The study area is based in the southern Caribbean region of the country near its border with Panama.This region is characterized by two dry seasons (February-March and September-October) and two rainy seasons (November-January and April-August).During the study period, the average monthly precipitation varied at around 234 mm, with an average temperature of 26.5 C, a maximum of 33.6 C and a minimum of 19.0 C (www.imn.ac.cr).
Vegas-Las Palmas parcels were established in 1984 as part of a rural settlement program of the government's Institute of Agrarian Development (IDA, by its acronym in Spanish).The settlement consists of 94 parcels within an area of approximately 300 hectares, intended for the cultivation of bananas, squash, maize, papayas, cocoa, rambutan, and Panamanian pepper.The community is bordered to the east by the Quiebra Caño (a primary canal that collects waters from the parcels and banana plantations and channels it into the Sixaola River), to the west by the town of Sixaola and the Finca Bananera Costa Rica, to the south by the river Sixaola, and to the north by the Finca Bananera Del Monte.The geology of the area consists of Quaternary alluvial formations, made up of clastic marine deposits, limestone, and clastic and volcanic rocks (Kapp ).The area is inhabited by 164 people in 43 households, of whom 56.4% are men and 43.6% are women.

Survey
A baseline study was carried out in the Vegas-Las Palmas parcels using a semi-structured instrument that gathered information about the basic services provided in the community, waste management, characteristics of wells, rainwater harvesting and farming activities.The instrument was applied to each of the 43 dwellings located within the parcels, and the results obtained served as input for the investigation of the well water quality used for human consumption.

Sampling and analysis
Water samples from 12 wells were collected during six

Data analysis
Data analysis was carried out using nonparametric survival procedures (Hensel ) with the NADA library of the R statistical package (Lopaka ).The maximum likelihood estimation (MLE) method was used to calculate the mean and standard deviation when the percentage of data below the detection limit (<QL) was less than 80%.The Kendall's tau correlation coefficient was calculated to evaluate the degree of association between water quality parameters.
Subsequently, a multivariate analysis was performed using ordinal nonparametric methods assigning tied ranks to the data below the detection limit.Analysis of similarities (ANOSIM) was used to evaluate the degree of similarity in water composition among wells and between seasons (999 permutations); and the degree of clustering of the sites sampled was established using a hierarchical cluster analysis (CA) using Ward's method (Ward ) and squared Euclidian distance as a measure of similarity.

RESULTS AND DISCUSSION
Despite the advances achieved in Costa Rica in terms of the quality, quantity and accessibility of the water supply, there are still marginalized communities that do not have potable water.In the case of the Vegas-Las Palmas parcels in Sixaola, Regarding sanitation, the parcels lack adequate wastewater management, where a latrine is the common system to treat human excreta (Figure 2 Concentrations of fluoride, chloride, nitrate and sulphate were below the regulations, spatially and temporally. Nevertheless, there may be an exposure issue for some ions such as nitrate which can cause methaemoglobinemia in infants when it is ingested in high amounts (WHO ).
The maximum nitrate concentration observed was 27.70 mg/L.In contrast, in the areas with the highest population density in Costa Rica, it does not exceed 5 mg/L (Reynolds-Vargas & Richter ).This situation shows the inequality related to access to good water quality that may also be present in other rural areas in the country.
Maximum values for Ca 2þ , Mg 2þ , Na þ and K þ in specific wells and samples also exceeded the regulations.The concentration of these ions is related to the lithology of the area and the rock-water interactions, but they can be increased by the dragging of materials in periods of extreme rainfall or floods.
Moreover, some trace metals exceeded reference standards for drinking water.The maximum concentration of Cd and Pb exceeded the recommended values in four and three of the sampled wells, respectively.These metals are mostly related to industrial waste but may also be related to the mismanagement of materials such as plastics, PVC, paints, pesticides or even corroded materials stored near the wells The maximum concentration of Mn exceeded the permitted limits in wells 1, 4, 6, 7, 8 and 12 (50% of wells sampled), reaching a maximum of 3.66 mg/L (Supplementary Table S4).S5).These results clearly indicate that water is not safe for human consumption.TDS and Na þ , hardness and Cl À , hardness and SO 4 2À , hardness and Mg 2þ , Cl À and Ca 2þ , Cl À and Na þ , Cl À and K þ , SO 4 2À and Na þ , SO 4 2À and Mg 2þ ; and Ca 2þ and Na þ .Finally, significant weak correlations (coefficient < 0.3) were found for pH and Cl À , pH and Fe, turbidity and Mn, TDS and K þ , TDS and Mn, TDS and faecal coliform, hardness and K þ , hardness and Na þ , hardness and Mn, F À and Zn, F À and Mn, F À and Pb, F À and Fe, Cl À and Mn, Cl À and Mg 2þ , Cl À and NO 3 À , NO 3 À and K þ , NO 3 À and Na þ , NO 3 À and Mn, Ca 2þ and Na þ , Ca 2þ and Mn, K þ and Cd, K þ and Na þ , Na þ and Mn, Mg 2þ and Pb, Zn and Fe, Zn and Mn, Zn and Pb, Zn and Cd, Fe and Mn, Fe and Cd, and Pb and faecal coliform.In general, conductivity, TDS and hardness showed positive correlations with the most abundant ions (Ca 2þ , K þ , Na þ , Mg 2þ , Cl À and SO 4 2À ), whereas trace metals such as Zn, Fe, Mn, Cd and Pb correlated with fluoride.The concentration of these latter species may be related to anthropogenic sources of pollution.
Water quality conditions were similar during the study period (Supplementary Table S2).No significant differences were found between seasons (R ¼ 0.015, p ¼ 0.312).However, no sample campaigns were carried out during the first months of each year or after flooding events.Inhabitants reported flooding events during certain months of the year, which carry high loads of contaminants and eventually affect water quality.These events are caused by the overflow of the Sixaola River and the irrigation channels, due to the heavy rainfall and the flat geomorphology of the terrain.After flooding events, wells cannot be used for long periods of time until they are cleaned and stabilized.
In these periods, inhabitants use mainly rainwater for drinking purposes.
The water composition among wells is very heterogeneous.There is a high degree of dissimilarity (R ¼ 0.5988, p ¼ 0.001), even when they are located relatively close to each other.The high spatial variation of some parameters can be observed in Figure 3.These variations reflect the degree of vulnerability of some wells due to the influence of meteorological, geological and anthropogenic (such as disposal of excreta and well conditions) variables in water quality.For instance, data variability may reflect the difference in depth, infrastructure, maintenance and proximity of potential sources of pollution of each well.
The well location in each dwelling is important due to their proximity to activities that could generate contamination, as previously mentioned.The daily productive and domestic activities can generate contaminants that can be transmitted by water and affect the health of the population.
For example, the wide use of fertilizers could increase nitrate concentration in water.In the area, there is an open aquifer where the phreatic level is located a few metres below the surface, which together with the characteristics of a permeable soil, allow the percolation of liquids and dissolved substances to easily pass through soil strata, allowing their contact with groundwater (Dey et al. ).
Finally, cluster analysis allowed the identification of three main groups with a D link /D max < 60 (Figure 4).The first cluster consists of wells 7, 10 and 11, which are those with the worst water quality conditions.These wells were characterized by high concentrations of turbidity, NO 3 À , Cl À , Fe, total coliforms and faecal coliforms.They are also more vulnerable to the impact generated by activities carried out in their surroundings due to deteriorating infrastructure.
The second cluster consists of six wells (3, 4, 5, 6, 8 and 9), which have generally lower concentrations of major ions, faecal coliforms and E. coli.This group supplies most of the population, and some wells have infrastructure that can help to prevent contamination.Finally, wells 1, 2 and 12 make up cluster 3, which had the lower microbiological contamination, despite the fact that some values for these wells (such as Ca 2þ , Mg 2þ , K þ , conductivity and turbidity) were also found to be outside the standard limits considered fit for human consumption.This information may be useful for the implementation of remediation strategies in the short term.For instance, inhabitants that use wells of cluster 1 are in a more vulnerable condition.It is important to highlight that with such high concentrations of coliforms, nobody should be using water from those wells for drinking purposes without adequate previous treatment.
Regardless of all the data presented above, especially the microbiological data, Costa Rica is among the Latin American countries with fewer waterborne protozoa parasites studies reported (Rosado-García et al. ).Nevertheless, specific information about waterborne diseases in the area is not always available, so the number of cases reported in the literature may be underestimated.On the other hand, Costa Rica is among the countries with better water  supply in the region.Only 0.5% of its population lack this service in comparison with the average of around 10% in Latin America and the Caribbean (Montgomery & Elimelech ).Our results indicate that special attention must be paid in rural areas, where data regarding water quality are scarce, but also that more extensive studies should be undertaken.This information can lead to a better coverage of safe and good quality drinking water in the country.

CONCLUSIONS
The importance of adequate water provision (quantity and quality) for public health cannot be overestimated.In the absence of adequate water provision for human consumption in the community of Vegas-Las Palmas parcels, its inhabitants are obliged to use other sources including wells, rainwater and river water.These sources do not ensure the quantity and quality of water needed to meet their basic needs.The data presented about well water were outside the standard regulations, which means it is not safe for human consumption.Attention should be paid to this situation since most of the population reported using this source.
The values of faecal coliforms and E. coli showed the absence of safe water in the area and may place the population at high risk of diseases caused by pathogenic microorganisms.The results obtained in this study show the importance of carrying out a more extensive study, including pesticides and risk assessments, to better estimate the impact of well water consumption on the health of the community.
Further investigations have to be developed in these areas.
Finally, financial investment to improve the existing wells' condition and/or the construction of an adequate distribution system that complies with national and international standards should be evaluated to ensure safe drinking water for the whole population.
sampling periods (n ¼ 72): June and October of 2014; April, September and November of 2015; and May of 2016.The wells were selected based on (1) that they were being used for drinking purposes, (2) a higher number of inhabitants were supplied from the same well, and (3) the spatial distribution of houses in the parcels.The samples for physical and chemical analyses were collected in high density polyethylene (HDPE) bottles previously washed with 3% m/v HCl and deionized water.Microbiological samples were collected in 100 mL sterile non-reusable vessels and stored separately to avoid contamination.Samples were immediately stored at 4 C and transported to the laboratory within 12 hours of collection.Physical, chemical and microbiological analyses were performed using the Standard Methods for the Examination of Water and Wastewater methodology (APHA/AWWA/WEF ).Temperature and pH were measured with an Oakton 300 multiparameter meter (IL, USA), and conductivity was measured with a Thermo Orion Star A222 meter (MA, USA); in both cases the measurements were in situ.Turbidity was measured with a Scientific Inc. Micro 100 turbidimeter (FL, USA) and water hardness was determined by titration with a EDTA standard solution.Gravimetric analysis at 180 C allowed the determination of total dissolved solids (TDS) previous water filtration through 0.45 μm pore filter (Advantec ® GC50).Atomic absorption spectroscopy (AAnalyst 800, Perkin Elmer, CT, USA) was used for the analysis of the most abundant cations and trace metals, where Ca, Na, K, Mg, Fe, Cu, Zn, Mn and Cd were analysed with an air-acetylene flame, while Pb, A, and Cr were analysed in a Zeeman-effect graphite furnace.F À , Cl À , NO 3 À and SO 4 2À were analysed by ion chromatography (ICS-90, Dionex, CA, USA).Finally, analyses of total coliforms, faecal coliforms and Escherichia coli were performed in the certified (INTE-ISO/IEC 17025:2005) San Martin Laboratory (San Jose, Costa Rica) using the multiple tube fermentation technique.

Figure 1 |
Figure 1 | Study area and sample sites.
Figure 2(a)).These shallow wells are dug by hand and provide access to water layers with depths ranging from 0.25 m to 5.12 m.Some of them do not have culverts or covers, which permits the entry of contaminants (Supplementary Figure S1, available with the online version of

Figure 4 |
Figure 4 | Box-plots of water quality indicators in Vegas-Las Palmas parcels, Sixaola.
The maximum conductivity value was 670 μS/cm, (c)).These systems are approximately 6 m in depth, lack a drainage system, and are sometimes at a higher altitude than the well, so leaching of pathogens may occur.Untreated wastewater is discharged directly in the soil in areas close to the wells; this practice is common in rural areas in Costa Rica.In fact, only 37% of the total wastewater discharges in the country are adequately treated (Valverde ).In addition, farmers keep animals, such as pigs, cows, horses, goats and chicken, whose excreta is not deposited in appropriate areas.Solid waste management is also an issue, where 72.1% of the population burns solid waste and 2.3% buries it without any separation.Most of these potential sources of pollution are less than 40 m from the wells (Supplementary TableS1, available online), failing to comply with the protection radius established in the national regulation.These situations represent a clear issue that may influence water quality in the area.characteristic of the climate in the area, and the sun's rays directly affect some wells because of the shallowness and the lack of infrastructure (such as culverts and covers).pH values below the regulations were found (ranging from 5.50 to 7.29); however, the average was within an acceptable range.

Table 1 |
Summary of the physical, chemical and microbiological parameters of wells in Vegas-Las Palmas parcels, Sixaola .Mena-Rivera & J. Quirós-Vega | Assessment of drinking water suitability in low income rural areas Journal of Water and Health | 16.3 | 2018 TDS, total dissolved solids; SD, standard deviation; WHO, World Health Organization; CR, Costa Rican legislation.aCalculatedusing maximum likelihood estimation method.Lfrom https://iwaponline.com/jwh/article-pdf/16/3/403/245913/jwh0160403.pdf by guest that is not destined for human consumption(Scott et al.