Assessment of Hydrological Properties and Proximate Impact of Septic Tank Leachate on Well-water Quality in Two Residential Areas in Ibadan, South-western Nigeria

Aims: To determine the impact of leachate from septic tank on proximate well-water quality in two different residential areas and the variation in the physico-chemical parameters of the well-water that are associated with spatial geographical location. Study Design: Randomized monthly collection and analysis of well-water over four months in two chosen residential areas. Methodology: The well-water samples were collected from 30 sites once every month; from different wells located at various perimeters from the septic tanks. The distance between the septic tanks and the wells were measured and the water subjected to physico-chemical analysis and bacteriological assessment to evaluate their qualitative, spatial and temporal variations. Results: A significant increase (p<0.01) was found in results from coliform counts between dry and wet seasons, while significant decrease (p<0.05) was recorded in the concentration of phosphate, salinity, total dissolved solids and potassium. The distance from the well to the septic tanks exhibited a negative correlation with coliform count (p<0.05), as well as for phosphate, nitrate, chlorine and ammonia (p>0.05). The Discriminant Chi square ( X 2 = 62.526, p<0.01) and Wilk’s Lambda (0.058) revealed a significant discrimination between the two study sites. Partial Eta Squared value of 0.740, 0.382 and 0.137 were reported for location, proximity of septic tank to well-water site and well depth respectively, showing their degree of contribution to variation in parameters measured. Conclusion: The results stressed the need to set standards concerning the distance and location of wells from septic tanks/septic tank, while considering spatial and temporal variations in hydrological environment of well-water sites.


INTRODUCTION
As population grows and urbanization increases, more water is required and greater demand would be placed on ground and surface water [1]. The rate of urbanization in Nigeria is alarming as the major cities are growing at rates between 10-15% per annum [2]. The human activities associated with such growth (including soil fertility remediation, indiscriminate refuse and waste disposal, and the use of septic tanks, soak-away pits and pit latrines) are on the increase, with concurrent increase in underground water pollution.
Groundwater contributes only 0.6% of the total water resources on earth, yet, it remains the major source of drinking water in Nigeria, as in every part of the world [3]. Ideally, potable water sources must be highly pure and free from chemical and microbial contamination. However, this useful resource is under threat of pollution from human activities as evident in low level of hygiene practices [4][5][6], and lack of recommended 'safe distance' between wellwater/borehole and septic tank sites in Nigeria. In view of this, monitoring the chemical, physical, and microbiological quality of groundwater is as important as assessing its quantity [7].
The two main contributors to the variations in the biogeochemical composition of groundwater are weathering processes and anthropogenic inputs [8]. Rapid increase in population, changes in life style, changes in living conditions and domestic waste management system [9] in Nigeria have resulted in a dramatic increase in the potential infiltration of extraneous materials into the groundwater environment [10]. In recent times, the impact of leachates on groundwater and other water sources has attracted a lot of attention because of its overwhelming environmental significance; hence, protection of groundwater is now an important environmental issue [11][12]. Leachates from septic tanks and other municipal sewage contain variety of chemicals like detergents, germicides, complex organic compounds and metals. Besides, human effluent composite and water contaminated with these sewage effluents may contain pathogenic organisms, high load of nitrate, phosphate, ammonia, total dissolve solid and associated complex organic substances which may be hazardous to human health if such water is consumed without treatment [13,14]. Additionally, uncontrolled microbial actions may result in the release of more toxic substances due to cumulative or synergistic effect of previously free or non-reactive component of the waste. Hence, the discharge of wastes from municipal sewers is one of the most important water quality issues worldwide and it is of particular significance to drinking water sourced from shallow water table [13,15]. During leachate percolation, water present in the waste and those generated by biodegradation serve as vehicle for leachate's vertical and horizontal migration finding its way into the groundwater environment thereby contaminating the groundwater. Survival and dispersal of microbes in the groundwater environment also varies, for instance Escherichia coli have been reported to have moved 46 m vertically and 70 m horizontally in an aquifer under favourable conditions, while enteroviruses have been reported in groundwater 183 m from a wastewater point source [15]. Most chemicals found in drinking water are of health concern, though only few of these chemical contaminants have been shown to demonstrate adverse health effects in humans as a consequence of prolonged exposure through consumption [16]. Consequently, evaluation of the microbial, physical and chemical quality of water is an important means of securing potable water for daily consumption. This study seeks to assess the proximate impact of septic tank leachates on the physico-chemical and microbial quality of groundwater in different hydrological settings viz: Agbowo (densely populated) and Akobo (sparsely populated) residential areas of Ibadan, Nigeria. This will provide a baseline information towards exploring a possible minimum distance that could be recommended for Nigeria, where well-water is the most common water source within the cities.

Location and Geology of the Study Area
This study was carried out in Agbowo and Akobo residential areas of Ibadan. Ibadan is located in south-western Nigeria, 160 km inland northeast of Atlantic coast of Africa, situated in-between rainforest and savanna. The study areas lie within latitudes 7°25'37"N and 7°26'26" N and Longitude 3°54'25" E and 3°56'50" E ( Fig. 1).
The climate of the study areas is tropically wet and dry, with the wet season from mid-March to October and dry period from November to March. The study was conducted within the Precambrian Basement Complex terrain of south-western Nigeria [17]. The main lithology of the rock unit include the amphibolites, migmatite gneisses, granites and pegmatites. Other important rock units are the schists, made up of biotite schist, quartzite schist talk-tremolite schist, and the muscovite schists [18].
The topography of the city is characterized by undulating terrain with general elevation between 180 m and 210 m above sea level, and is drained by rivers Omi, among others. Occurrence of groundwater is dependent on the extent of weathered units and development of secondary porosity, e.g. fractures, faults, joints [19]. Consequently, the weathered aquifers are generally discontinuous, with groundwater occurrences in localized disconnected phreatic weathered regolith aquifers, essentially under unconfined to semi-confined conditions [20].

Sample Collection
This study was carried out between April and August, 2012, across wet and dry seasons. Wellwater samples were collected from thirty-one locations in the study site. Hand held Global Positioning System (GPS) receiver device with 12-channel tracking and differential correction capability was used to determine and identify location where the water samples were collected.

Quality Assurance Procedures
Samples were collected in plastics that were prewashed with detergent water solution, rinsed with distilled water and soaked for 48 h in 50% HNO3, then rinsed thoroughly with distilled deionised water before air-drying. As part of the quality control measures, samples' bottles were rinsed with sample well-water before filling. Samples for Calcium (Ca were refrigerated at 4°C prior to analysis.

Well Depth and Water Level Determination
Well depths were determined using graduated tape with added weight and a similar procedure was employed for water level determination.

Determination of Physical and Chemical Parameters
Non-conservable parameters like, pH, Salinity and other parameters like Total Dissolved Solid (TDS), conductivity, temperature (°C), Dissolve Oxygen (DO) and percentage Oxygen (%O 2 ) were taken immediately at the sampling point using Consort CS-C933 Multi-Parameters Portable meter, Topac Instrument, Inc. USA. Total hardiness, Calcium (Ca) and Magnesium were determined by Ethylene Diamine Tetraacetic Acid (EDTA) method. Flame photometric method was employed in the determination of Sodium (Na + ) and Potassium (K + ); Chloride (Cl -) was measured by potassium chromate method. Nitrate (NO -3 ) by Phenoldisulphonic acid method, Phosphate (PO 3 -4 ) by Ascorbic Acid Method and Ammonia by Indophenols (Colorimetric).

Total Coliform Assessment
Sterile corning tubes were used to collect wellwater samples from various sampling sites and were preserved at 4°C prior to microbial analyses. Microbial analyses were carried out within 24 hours of collection. Borosilicate glassware were used in microbiological testing in conformity with Section 9000 of "Standard Methods for the Examination of Water and Wastewater". Dilution bottles were made of resistant Borosilicate glass, with plastic screw caps which were equipped with liners that do not produce toxic or inhibitory compounds when sterilized. McConkey broth with incorporated Bromocresol purple as indicator (LAB M Ltd. UK) was used as the growth media. Sterile syringes were used to introduce 5 mL of the broth media (3.5% w/v) into 160/15 mm test-tube and small Durham tube inverted inside the test tube (totally submerge) to determine gas production in the sterilized fermentation tube. Series of 100%, 10% 1% and 0.1% of the stock solution of wellwater samples (1 mL) were prepared with distilled water. The prepared concentrations were introduced into sterilized fermentation tubes.
Most Probable Number (MPN)/100 mL were estimated using MPN Index of Standard methods 9221 Standard Total Coliform fermentation technique [21]. A multiple-tube fermentation technique was used for total coliforms (McConkey Broth Medium) and various combinations of positive results from five tubes per dilution were used (10 ml, 1.0 ml and 0.l ml sample portions).

Data Analysis
The bacteriological data were log-transformed to meet the assumption of equal variance which were then subjected to student T-test to evaluate the impact of seasonal variation on coliform density. Correlational analysis were performed to evaluate the extent of the relationship between coliform density, well depth and distance between well and the nearest septic tank point for dry and wet season. Correlation analysis was performed on the parameters measured to evaluate their inter-relatedness. Principal Component Analysis (PCA) was also used to assess the concentration of physicochemical and bacteriological data from well-water samples, quantitatively determining the minimum number of new variables necessary to reproduce various attributes of the data. Discriminant analysis was performed on the parameters to identify variables that best differentiate the two sampling zones (Agbowo and Akobo). The Water Quality Index (WQI) was determined using the Weighted Arithmetic Index method [22].

Variable Analysis of Well-water Samples
The descriptive statistical analysis data presented in Table 1 show the univariate summary of well-water sample analysed during the study period between the month of April and Generally, underground water is believed to be the purest source of water [23] because of the purification properties of soil [24]. However, the amount and variation in the level of dissolved materials in underground water is a complex function of interaction between climate, land use patterns, human activity, weathering and the geologic make-up of the hydrologic environment. These functions vary from one geographical area to the other [25]. The result presented in Table 1 shows that the overall water hardness, DO, coliform, distance from nearest septic tank, salinity, TDS, conductivity, ammonia, nitrate, phosphate cluster less around the mean, while, temperature, pH, potassium, calcium, well depth and Percentage Oxygen clustered more. The degree of variation observed in the data structure is in agreement with previous studies [26][27]. A wide variation was observed in the physicochemical variables of well-water collected from Agbowo and Akobo as shown in Table 1

Coliform Count in the Dry and Rainy Seasons
The presence of coliform was confirmed by lactose fermentation and concurrent gas production. There was a strong negative correlation (r = -0.51, P < 0.05) between coliform count and distance from septic tank for the dry season (Table 2) and the strength of correlation coefficient increased during the raining season (r = -0.58, P < 0.05) compared to dry season (r = -0.51, P < 0.01).
The correlation between coliform count and well depth was not significant (r = -0.16, P > 0.05) but there was an indication of a negative relationship between coliform count and that of well depth in wet season (r= -0.16, P>0.05). On the other hand, slightly positive but insignificant correlation (r = 0.1, P>0.05) existed between coliform count and well depth during dry season (Table 2). Well-water in Akobo study area recorded a low mean coliform count of MPN 674.14/100 mL with higher kurtosis (-1.85) compared with Agbowo which recorded a mean of MPN 1008.00/100 mL and -1.20 for kurtosis indicating that well samples coliform count in Akobo clustered less around high MPN value than Agbowo samples. The result of the microbial analysis shows high levels of coliform contamination in well-water collected in Agbowo. The low well-depth range recorded in Agbowo could account for the high level of groundwater contamination as against what recorded in Akobo. The low sanitation level and hygiene practices [28] and other environmental factors observed in the community [1] may have contributed to the high microbial assessment of Agbowo wells. It was observed that the layout of the buildings in Agbowo were highly clustered and not well planned, resulting in the close distance of well sites to septic tanks and refuse dumps. An extreme observation involved cases in fenced neighbourhoods where wells were ignorantly sited at close proximity to the septic tank of the neighbouring house. This might have affected the reported high value of Coliform density observed in site AG2 with > MPN1600/100 mL and it supports the report on human waste contamination of groundwater by USEPA [29].
The negative correlation obtained between total coliform count and distance from the nearest septic tank and well-depth (Table 2) suggests that the total coliform decrease with increasing well distance and well depth. This is in agreement with the findings of Adekunle et al. [30] and Adetunji and Odetokun [1] and also in agreement with the fact that soil filtration potential increased with distance both vertically and horizontally. However, the vulnerability of groundwater to particulate pollution is a function of the ease with which particulate and dissolve solutes can move with water and the attenuation capacity of the intervening materials [31]. The filtering capacity of soil as intervening materials will therefore depend on the soil properties and the distance between pollutant source and receiving water and also the properties of the pollutant. Previous studies [31] revealed that coliform could travel distance of 70.7 m from sewage trenches intersecting groundwater while covering a distance of 30.5 m within 35 hrs when travelling through sand and pea gravel aquifer. Francy et al. [32] also confirm an association between proximity of septic tank to well water site, well depth and coliform density.

Variation in Well Water Physicochemical Properties
A paired sample Student T-test (Two tailed) comparing mean differences that exist in microbial and physico-chemical parameters of well-water sample in the dry season (April) and wet season (Average of value obtained between May and August) was determined ( Table 3). The result shows that there was significant increase of coliform count between the dry season and wet season. (T = -3.401, P < 0.01) and pH (T= -3.566, P < 0.01), while there was a significant decrease in the concentration of phosphate (p < 0.01), salinity (P < 0.05), TDS (P < 0.01), potassium (P < 0.01) and sodium (P <0.01), conductivity (P <0.01) and ammonia (P <0.01).
The result of monthly variation observed in salinity, TDS and conductivity were as shown in Figs. 2, 3 and 4 respectively. There was a significant difference in Salinity (t = 8.234, df = 29, P <0.01) between samples collected from Agbowo to that of Akobo. The salinity of wellwater in Agbowo increased slightly from April to May and later declined through June and July and a slight increase in salinity was later observed in August. In Akobo, there was no obvious fluctuation in the salinity of the wellwater (Fig. 2). A significant difference was also observed in the TDS (t = 8.175, df = 29, P< 0.01) and conductivity (t = 8.206, df = 29, P <0.01) between Agbowo and Akobo study areas. The monthly variation in TDS and conductivity of Agbowo study area followed a similar pattern as that of Salinity. However, TDS and conductivity of well water in Akobo showed a slight decrease from the month of April through July before gaining stability (Figs. 3 and 4).
The increased concentration of Salinity, TDS and Conductivity between the month of April to M as shown in Figs. 2, 3, and 4 may have resulted from precipitation enhancing leaching of long term accumulated ionic and non-ionic particles (during dry season) on surface soil [26]. Precipitation also enhanced dissolution of materials naturally present in soil and the introduced water dissolved ionic and non minerals. The resulting water that moves through the soil to the underlying groundwater will then 7 from the month of April through July before The increased concentration of Salinity, TDS and Conductivity between the month of April to May . 2, 3, and 4 may have resulted from precipitation enhancing leaching of long ionic particles (during dry season) on surface soil [26]. Precipitation also enhanced dissolution of in soil and the introduced water dissolved ionic and non-ionic minerals. The resulting water that moves through the soil to the underlying groundwater will then be enriched with high load of soluble salt [25] and consequently enhance the leaching of septi tank scum into the groundwater environment. The subsequent reduction in salinity, TDS and Conductivity, may have resulted from increased dilution from continuous rainfall during the raining season [26] compounded by human withdrawal of previous solute laden water for domestic use.
The trend in variation of average Salinity, TDS and Conductivity was taken as the prime parameter to focus on monthly variations [33] for Agbowo and Akobo study areas.

Inter-relationships of Physical and Chemical Parameters
The result of the correlation analysis performed to evaluate the extent of the interrelationship that exists among the assessed variables is shown in Table 4. The correlation matrix revealed that strong positive relationship (P<0.01) existed between phosphate and the following parameters; nitrate, ammonia, chloride, sodium, magnesium, calcium, conductivity, TDS, salinity, and hardness (Table 4). In addition, strong positive correlation (P<0.01) also existed between nitrate and the following; ammonia, chloride, sodium, magnesium, calcium, conductivity, TDS and salinity.
Conductivity and TDS were perfectly correlated (r = 1.00) ( Table 4). A significant positiv correlation (P<0.05) was found between log transformed coliform count and other parameters

relationships of Microbial, Parameters
The result of the correlation analysis performed to evaluate the extent of the interrelationship that exists among the assessed variables is shown in 4. The correlation matrix revealed that relationship (P<0.01) existed between phosphate and the following parameters; nitrate, ammonia, chloride, sodium, magnesium, calcium, conductivity, TDS, salinity, 4). In addition, strong positive correlation (P<0.01) also existed en nitrate and the following; ammonia, chloride, sodium, magnesium, calcium, Conductivity and TDS were perfectly correlated 4). A significant positive 0.05) was found between log d coliform count and other parameters like phosphate, nitrate, ammonia, sodium, conductivity, TDS and salinity. Well distance from septic tank had a significant negative correlation (P<0.05) with average coliform count, and negative insignificant correlation (P>0.05) with phosphate, nitrate and ammonia ( depth recorded a significant negative correlation (P<0.05) with parameters like phosphate, nitrate, ammonia, chloride, sodium, TDS, salinity and hardness (Table 4a & The positive correlation observed between Coliform count, Phosphate, Nitrate, Chloride and Ammonia in ground water, was in line with previous work that had associated leachate from septic tank impacting on the concentrations of these variables concurrently [34 presence of coliform in the groundwater had been reported to indicate contamination by human and animal faeces [22,36]. These pathogens may pose a health hazard to infants , 2015; Article no. BJAST.12452 (TDS) of well water samples in of well water samples of Akobo and Agbowo like phosphate, nitrate, ammonia, sodium, conductivity, TDS and salinity. Well distance from septic tank had a significant negative correlation (P<0.05) with average coliform count, and on (P>0.05) with phosphate, nitrate and ammonia (Table 4). Well ficant negative correlation 0.05) with parameters like phosphate, nitrate, ammonia, chloride, sodium, TDS, salinity and ion observed between Coliform count, Phosphate, Nitrate, Chloride and Ammonia in ground water, was in line with previous work that had associated leachate from septic tank impacting on the concentrations of these variables concurrently [34][35]. The e of coliform in the groundwater had been reported to indicate contamination by human and animal faeces [22,36]. These pathogens may pose a health hazard to infants and people with weak immune systems [36]. The significant variations in coliform count during wet and dry seasons; in relation to distance of well from septic tank and well depth show that run-off and weathering resulting from precipitation during raining season had a positive impact on the migration of coliform from different pollution sources to the receiving underground water.
The strong co-linearity that existed between salinity, TDS and conductivity as expressed in the correlation matrix shows that salinity as a measure of the amount of dissolved particles and ion in well-water sample can account for TDS (measure of all dissolve substances) and conductivity (measure of charged ion or ionic particles) of the water sample. The negative correlation obtained between well distance from septic tank and soluble particles represented by salinity, TDS, conductivity show a possible contamination through leaching from proximate septic tank. This was justified by SWRC [25] findings that discharges from septic systems increased the salinity of the receiving water body. Most salt resulting from these soluble particles do not naturally degrade, and can remain in groundwater for decades [25].
The weak negative correlation that existed between well proximity to septic tank and Nitrate; phosphate and ammonia concentrations may suggest that leachate from septic tank contributed to the increased concentrations of these parameters [37][38][39]. The strong negative correlation reported for nitrate concentration and well depth agrees with Piskin [40] whose findings indicates that Nitrate concentration decrease with increase in well-depth. Nitrate and Ammonia are by-products of natural break-down of Nitrogen containing organic compounds from human and domestic wastes that accumulate in the septic tank and this is reflected in the observed strong positive correlation between the above two variables.
The negative correlation that was found to exist between well depth and the following variables: phosphate; nitrate; ammonia; chloride; sodium; magnesium; calcium; conductivity; TDS; salinity and hardness further establishes the fact that areas with high aquifer levels are most predisposed to particulate contamination [40]. This may account for the reason why 73.9% of well samples at Agbowo study area, with well depth range (0.25 to 5.5 m), were unsuitable for drinking; when compared to Akobo which recorded a well depth range of 4 m to 15 m with a lower percentage (54.55%) of unsuitable drinking water (Table 9).

Discriminant Analysis (DA)
The result of a one-way ANOVA discriminant analysis for the variables in Table 5 shows that temperature DO, pH, K, %O 2 and coliform count failed the tolerance test at 0.001 and did not contribute to the discriminant model classifying the two study area. The Eigenvalue value (16.151) for the first discriminant function accounts for 100% of the variance that exists among the variables obtained from the two study areas and the Chi square test (X 2 = 62.526, df=30, P<0.01) shows that the difference in the two groups was not just due to chance. The strong canonical correlation of 0.970, low Wilk's Lambda value of 0.058 shows a strong relationship between the discriminant score and the group classification. Predicted group membership (Table 6) shows 100% perfect classification of sampling site based on discrimination between variables observed.
The low Wilk's Lambda value (0.058) reveals the strength of the variable at discriminating between the two study areas in decreasing order: phosphate > nitrate > salinity > conductivity >hardness>TDS > chloride > ammonia > magnesium > sodium > calcium. The standardized coefficients comparing variables measured on different scales downgraded the discriminative ability of phosphate to 5th position while promoting discriminant ability of conductivity, sodium, TDS and hardness above phosphate (Table 5).
Physical parameters like temperature, DO, and %O 2 and coliform count did not contribute to the classification of the study areas. It may be that geographical location with respect to groundwater hydrological properties did not have an influence on the variation in the physical and microbial properties of the groundwater between the two study sites.
The result of the discriminant analysis (Tables 5 and 6) shows that differences which exist among variables in the study areas were not only limited to the proximity to septic tank or water level nor due to chance, therefore the significant difference in the hydrologic properties of the two locations might be a contributory factor [41][42].

Regression Analysis for Multiple Dependent Variables by General Linear Model (GLM)
The result of the GLM Multivariate procedure presented in Table 7 provides regression analysis for multiple dependent variables by factor variables (location of the study area) and the covariates (well closeness to septic tank and well depth) based on the model, in which location, distance and well depth were assumed to have linear relationships to the physicochemical parameters and coliform count of the well water.  (Table 7), showing their degrees of contribution to the model in that order. The multivariate GLM presented in Table 7 conclusively reveals that location contributes more to the variation that exist among physical and chemical parameters of the well water examined, than septic tank proximity to well site and well depth. However, proximity to septic tank has a greater impact on the coliform count. Most of the well water quality in Akobo and

Water Quality Index (WQI)
All the physico-chemical and Microbial parameters analysed were used to calculate the WQI in accordance with the procedure explained in the methodology section. The results of the WQI were presented in Tables 8 and 9. Only 16.13% of the sampled well water fall within the excellent water quality range and 9.68% were within the very poor water range; while 74.19% of the sampled well waters were unsuitable for drinking. The quality of the well water collected from the two different study areas have shown that in Agbowo, 73.9% of the water samples collected were unsuitable for drinking, while 54.55% of that collected from Akobo were unsuitable for drinking. High coliform count was the predominant variable responsible for the poor quality of the water sample in Akobo, while high chlorine and coliform count were responsible for the poor well water quality in Agbowo.    Nitrate level exceeding the recommended standard (45 mg/L) as observed in Wells AG4, AG12, AG13 and AG15 sampled and this has been reported to cause methemoglobinemia in infants [35,36,[44][45], thereby rendering the water in these wells unsafe for infant consumption.
High level of Chloride observed in all Agbowo sample site (Average 299 mg/L against 200 mg/L NESREA Standard) has been associated with sewage pollution [22]. Chloride is often chemically bonded with sodium naturally present in soil to form Sodium Chloride which impact salty taste on water. The simultaneous fluctuation in the concentration of some variables may be accounted for by infiltration of rainwater through the porous and permeable unconfined aquifer into shallow water table that enriches the groundwater with dissolved solute taken from surface and subsurface soil. From observation of water conductivity, TDS and salinity decreased immediately after rainfall.
Slightly Low pH level in AK5, AK7 and AK10 well water (below recommended standard 6.5 to 8.5)) observed in some sample points in Akobo may be accounted for by the low level of calcium and magnesium concentration in the study area [46].

CONCLUSION
In this study, various statistical techniques were used to evaluate quality and variation in physical, chemical and microbial parameters of well-water (underground water) in Agbowo and Akobo in relation to their proximity to septic tank. The results have shown that leachates from septic tank owing to proximity to well-water impact on physic-chemical properties such as nitrate, Ammonia, chloride, Sodium, Magnesium, Conductivities, TDS, Salinity, water hardness, and microbial qualities of the well-water. Precipitation also contributes significantly to the migration of coliforms from various sources to the underground water.
There is therefore a need for biogeochemistry assessment of hydrologic environment before considering siting a well in any location. The results suggest that the minimum standard distance between septic tank and well water will vary with different hydrologic environments and recommend a restriction on digging wells for domestic use in areas of high aquifer levels; rather, boreholes should be an alternative for the provision of potable water in such areas.