Introduction

Groundwater is one of the principal sources of water for meeting the domestic as well as agricultural requirements in India. The freshwater demand has increased in last few decades with increase in population. Traditional source of freshwater for all season is groundwater and it is being used for agricultural, industrial and daily needs (DDWS 2009; World Bank 2010). Rapid industrialisation, urbanisation and extensive use in agricultural sector have rendered groundwater resources vulnerable to depletion and degradation in quality. Groundwater source is the most superior in all forms of water sources due to its natural filtering effect and suitability of groundwater can be determined by assessing groundwater quality. Many researchers (Vzquez-Su et al. 2005; Hassane et al. 2016; Yang et al. 2017) stated that the groundwater chemistry is governed by natural processes (hydrogeological conditions, soil-rock-water interaction etc.) and anthropogenic activities (industrial, agricultural waste disposal, sanitation disposal). Among the anthropogenic stresses, on-site sanitation system has emerged as polluting source of groundwater and its effect is observed on key parameters like NO3, Cl, Total Dissolved Solids (TDS) and faecal coliform (ARGOSS 2001). The groundwater sources near to the on-site sanitation disposal tank are highly vulnerable to contamination (WHO 1993).

India has been attempting to tackle open defecation by providing on-site sanitation and off-site sanitation. However, the rapid urbanisation is challenging to the governmental bodies to provide clean water and sanitation facilities (UNICEF 2002). Densely slum areas preferred on-site sanitation system (Fig. 1) since the off-site sanitation and on-site sewerage treatment require significant capital investment. As compared to other metropolitan cities in India, coastal city Chennai has the highest sewerage network coverage (94%), but the coastal slum area adopted on-site sanitation system to dump its waste. Many authors have proved that in coastal region aquifers, groundwater quality is more vulnerable to sew water intrusion as well as impacts of anthropogenic activities (Hagedon 1984; Currell et al. 2015; Bhosale and Kumar 2002; Soni and Pujari 2012; Gopinath et al. 2016; Dhiman and Thambi 2009). Some researchers (S Sathish et al. 2011; Indu Nair et al. 2013; Indu Nair et al. 2015) have reported seawater intrusion in Chennai area. Present study has selected two (Urur Kuppam and SKP Puram) such sites near the coast of Chennai city where on-site sanitation system is adopted.

Fig. 1
figure 1

On-site sanitation system toilet used in the experimental area

Full size image

Most of the previous on-site sanitation studies (Taru and WEDC 2005; Pujari et al. 2007; Pujari et al. 2011; Jangam et al. 2015) in India are done in hard rock terrain and in alluvial settings but there is need to study the coastal aquifer where groundwater quality is affected by complex sources such as on-site sanitation, and seawater intrusion. The aim and purpose of this study is to monitor the groundwater sources and identify the possible impact of on-site sanitation system on coastal groundwater quality.

Study area

The Chennai city is located on the southeast coast of the Indian peninsular and bounded by longitude 80° 17′ E and latitude 13° 04′ N. The present study has been carried out in south Chennai coastal aquifer (Fig. 2). The city gets annual average rainfall 1200 mm chiefly a contribution of a southwest monsoon from July to September and northeast monsoon during October to December (Ganasundar and Elango 1998). The city has a mean annual temperature 24.3 to 32.9 °C with high humidity range 65 to 85%. Regional topography gradients towards the eastern side with undulating sand bars. The area is drained by Adyar river. The Chennai city has various types of geological formation from ancient Archaeans to recent alluvium. The Archaeans made up of charnockite are at the basement and are overlain by recent alluvium at the top. The thickness of alluvial cover increases from west to east as the result of Adyar river disposition. The major aquifer system is formed by this alluvial soil and fractured charnockite (CGWB 2008). Above the alluvial unconsolidated coarse grain sand, gravels, pebbles, clay and sandstone are found. This quaternary sediment formation constitutes shallow unconfined aquifer with annual water level fluctuation from 1.15to 7.93 m bgl in summer and 0.15 to 5.63 m bgl in monsoon. The major water-bearing formation is the coastal sand. Most of the wells are tapping the top quaternary aquifer. Indiscriminate pumping of groundwater along the Chennai coast led this aquifer under constant threat of salt-water intrusion.

Fig. 2
figure 2

Study area and sampling locations

Full size image

Two different sites namely Urur Kuppam (site 1) and SKP Puram (site 2) were selected for the present study where on-site sanitation system was adopted. Both study locations are situated on the bank of the Adyar river covering approximately 2 sq. km area. Urur Kuppam (site 1) is a densely populated slum area having nearly 500 hut houses with 1200–1500 population whereas SKP Puram having 150 houses with population around 500. Satpaty et al. (Satpathy et al. 1987) observed the mixing of seawater in the Adyar river in monsoon season. The groundwater sources in Urur Kuppam (Fig. 3) are scattered around the community toilet whereas in SKP Puram (Fig. 4), it is scattered around individual septic tanks. Site 1 is slum area with temporary hut houses and some bricks houses. The area has only a community toilet run by an NGO. The groundwater samples (8 nos. from each site) were collected from bore well tapping from the top aquifer at shallow depth.

Fig. 3
figure 3

Sampling location in Urur Kuppam (site 1), Chennai city

Full size image
Fig. 4
figure 4

Sampling location in SKP Puram (site 2), Chennai city

Full size image

Methodology

A detailed field study has been conducted in the year 2000. The sampling was conducted in the months of May and August for monsoon. All sampling locations from both the sides are bore wells which are tapping the top aquifer. The depth of the bore well is ranging from 4 to 12 m. Extensive care has taken during sampling, preservation, handling and analysis of the samples. All the sampling bottles and equipment are acid-rinsed and cleaned twice with double-distilled water before use. Pre-cleaned 1000-ml polyethylene bottles were used for collection of samples for physico-chemical analysis. Physical parameters namely temperature, pH and electrical conductivity (E.C.) were measured using portable sensor from Eutech Instruments. Bacteriological samples were collected in pre-cleaned sterilised wide-mouth 300-ml borosilicate glass bottles. The bacteriological samples were shifted to the laboratory for further analysis at freezer condition (4 °C). Before analysis in the laboratory, all the groundwater samples were filtered through Whatman filter paper (no. 4). The physico-chemical parameter was analysed by following standard protocol (APHA 1998). Sodium as Na+, potassium as K+ and calcium as Ca+ were analysed using a flame photometer (model–CL361), with 0.5, 0.5 and 15 ppm, respectively. The concentration of nitrate as NO3− was detected using UV-visible spectrophotometer 118 (Make-systronic). Chloride as Cl− was determined by titration method. Bacteriological samples were examined using membrane filtration technique (APHA 1998). For FC determination, colony formation was done by taking 50 ml of sample. The bacterial contamination was measured in colony formation in (CFU/100 ml) after incubating sample at 45.5 °C for 24 h.

Results and discussions

Hydrochemical facies

The samples were analysed for physico-chemical as well as microbial analysis. The statistical analysis like minimum, maximum, mean, median and standard deviation (SD) of key parameter data of both sites at different seasons is summarised in Table 1. All samples are compared with the Bureau of Indian Standards (BIS) of drinking water guidelines (BIS 1991). The concentration of TDS in groundwater samples collected at both sites is shown in Fig. 5. The average TDS concentration found at site 1 (3324.3 mg/l) is less than the average concentration at site 2 (5671.9 mg/l) in both seasons. Highest TDS concentration (9960 mg/l) was found at Urur Kuppam (site 1) in monsoon season. For TDS concentration, site 1 exceeds BIS limit for 5 nos. of samples in each season while at site 2, all 8nos samples for summer and 7 samples for monsoon are exceeding the BIS limit (2000 mg/l) (Fig. 5). The lower concentration of TDS was observed at site 1 in sample no. CO2 and CO3 which indicates the availability of fresh water. As a direct measure of salinity, TDS shows significant high values (9660 mg/l) in some observation wells and high chloride is also observed (6600 mg/l). The high values of TDS and chloride can be linked to the effect of seawater intrusion.

Table 1 Statistical hydrogeochemical data of the sampling sites at different seasons
Full size table
Fig. 5
figure 5

Seasonal variation in concentration of TDS at site 1 and site 2

Full size image

Groundwater sources in the vicinity of on-site sanitation system are expected to have higher concentrations of chloride and nitrate as these are the key chemical contaminants derived from the on-site sanitation. Human excreta can release 4 kg/year/person of nitrate from faecal matter and 4 g/day chloride concentration (Lawrence et al. 2001). Continuous loading of these contaminants can lead to groundwater pollution from on-site sanitation system. In this study, consideration of chloride as a key parameter is very difficult as the area has been affected by seawater intrusion. Figure 6 indicates the chloride concentration in the study area at different seasons. It is evident that the average chloride concentration at site 2 (2712 mg/l) is more than the average concentration at site 1(1871 mg/l). Most of the samples are exceeding the BIS limit for chloride concentration. The average value of chloride is observed more in site 2 than in site 1 in both seasons. The higher chloride concentration can be the results of seawater intrusion as higher levels of Na+ and Cl− are indicators for significant effect of seawater mixing (Mondal et al. 2008). It is very difficult to identify Cl− source, as the seawater intrusion has Cl− as a major contributor and sanitation impact analysis has Cl− as its key parameter. The nitrate concentration varies between 6–9 mg/l in summer and 47–89 mg/l in monsoon at site 1 and 1–28 mg/l in summer and 5–25 mg/l in monsoon at site 2 (Fig. 7). It is observed that all samples collected from site 2 do not have nitrate concentration exceeding the BIS limit (45 mg/l) for any season. A close examination of site 1 reveals that in summer all samples are below the BIS limit but exceed in monsoon for all locations. The high concentration of nitrate is often attributed to densely populated area (J. R. Fianko et al. 2009). The pattern of increase in nitrate pattern in monsoon and post-monsoon has been observed in previous studies (Pujari et al. 2007, Pujari et al. 2011). The low dissolved oxygen concentration in deep water table conditions in summer season creates anaerobic conditions and leads to conversion of nitrate to nitrogen gas (NEERI 2005). The higher concentration in monsoon can be a result of higher loading of unregulated sewage as well as the effect of landfill.

Fig. 6
figure 6

Seasonal variation in concentration of chloride at site 1 and site 2

Full size image
Fig. 7
figure 7

Seasonal variation in concentration of nitrate at site 1 and site 2

Full size image

Correlation matrix

Statistical correlation (Table 2 and Table 3) has been carried out for the parameters namely nitrate, chloride faecal coliform TDS and the distance of groundwater source from the septic tank. The results indicate that distance has positive correlation with faecal coliform at site 1 in summer (0.3027) and monsoon season (0.2122) as well as at site 2 in summer (0.35488) and monsoon (0.40064). In summer, only faecal coliform shows the positive correlation with distance while all other parameters are negatively correlated. It is observed that chloride has positive correlation with TDS in both seasons at both sites. The higher correlativity of chloride with TDS in both seasons indicates the seawater interaction. In monsoon, the parameters namely nitrate and faecal coliform show positive correlation with distance and with each other. The positive correlation of these three factors attributes the impact of another factor like same source. It is expected that there should be a negative correlation between distance and the faecal coliform concentration. In the present case, it is found to be positive which indicates that the contamination is dependent on other factors apart from the distance. Though there is negative correlation between distance and nitrate in summer season, the opposite is observed in monsoon. In monsoon, the positive correlation indicates that there may be other factors besides distance which may be contributing to nitrate concentration.

Table 2 Correlation matrix between the key parameters and distance between pit and sampling well in both seasons at site 1
Full size table
Table 3 Correlation matrix between the key parameters and distance between pit and sampling well in both seasons at site 2
Full size table

Hydrogeochemical facies

The evaluation of groundwater and chemical properties can be assessed using hydrogeochemical facies. More accurately, it can be understood by plotting cations and anions on a piper trilinear diagram (Mondal et al. 2008). The hydrogeochemical facies (Fig. 8) indicated that all the samples are mostly dominated by Na+, Cl− and SO4− and these are known representative ions of seawater. The dominant water type Na–Cl indicates the mixing of seawater with fresh water in aquifer. Gibbs plot was plotted to identify the mechanism controlling the groundwater and its chemistry (Gibbs 1970). The samples (16 nos.) were plotted for both seasons to identify the seasonal impact and found evaporation dominance. Figure 9a, b clearly indicates evaporation dominance controlling the groundwater chemistry. The degree of salinization in groundwater due to seawater mixing is indicated by the increasing ratio of cations and inions. The hydrogeochemical process and anthropogenic impact evident complexity of groundwater and need of sanitation impact study more accurate.

Fig. 8
figure 8

Piper plots showing hydrogeochemical facies

Full size image
Fig. 9
figure 9

Gibbs plot showing the mechanism of groundwater chemistry for seasons a summer and b monsoon

Full size image

Sanitation influence

Considerable variation in water quality and seawater mixing makes it difficult to identify the sanitation influence from its key chemical parameters. Previous studies on on-site sanitation indicate percolation of bacteria in top aquifer (Pujari et al. 2007) or the retention of the bacteria up to few meters after percolation (Jangam et al. 2015). Usually, the percolation of bacteria depends upon soil characteristics where on-site sanitation system has implemented as well as the nature of porosity in the aquifer system. As the city has a semi-arid climate and the aquifer depth is very shallow (4.5–8 m) with top cover of alluvial soil and sand, the survival rate of bacteria is favourable. It is observed that the concentration of faecal coliform is much more in monsoon than the summer season (Fig. 10). It may be due to the infiltration and recharge in the monsoon which can lead the faster movement in the vadose zone. Faecal coliform concentration at site 1 in the monsoon season is increased at all samples while for site 2 it is observed that only one location has more concentration in the monsoon season. Samples like CO1 and CO2 (site 1) which are close to the septic tank have lesser concentration as compared to the other samples. It is likely that the potential for migration in case of CO1 and CO2 may be less as compared to other locations.

Fig. 10
figure 10

Seasonal variation in concentration of faecal coliform at site 1 and site 2

Full size image

Nitrate in groundwater originates from the oxidation of ammonium in unsaturated zone (Zilberbrand et al. 2001). The source of ammonium is mainly from the disposed excreta into the top layer where ammonium concentration is dominant among the nitrogen compounds. The sources of ammonia in highly urban areas are domestic sewerage leaching, on-site sanitation etc. (Jacks et al. 1999). Groundwater affected wells by on-site sanitation are generally high in nitrate concentration (Pujari et al. 2007; Pujari et al. 2011). The marine influence is characterised by low nitrate and high chloride concentration (Appelo and Postma 2005) and it is amply evident in most of the samples collected from both sites (Fig. 11). Most of the samples are observed as chloride dominant while the samples influenced by the on-site sanitation practices are higher in nitrate concentration.

Fig. 11
figure 11

Chloride vs nitrate graph showing influence of origin

Full size image

Conclusions

The result of piper trilinear diagram and Gibb’s plot indicates that the groundwater chemistry of coastal aquifer is highly influenced by seawater–groundwater mixing. The accurate understanding of the impact of on-site sanitation on groundwater in coastal areas can be only studied by understanding the complexity of hydrochemical process. Groundwater contamination in areas served by on-site sanitation system is totally dependent on local hydrogeological characteristics like the nature of aquifer systems (Pujari et al. 2007, Pujari et al. 2011 and Jangam et al. 2015). The present study is focussed on analysis of the groundwater samples for physico-chemical and bacteriological parameters and assessing the contamination due to on-site sanitation systems. It is observed that the water quality in the present study is governed by the seawater intrusion as well as impact of on-site sanitation system. The impact of evaporation effect is also observed in the groundwater chemistry.

Recommendations

In the absence of centralised water supply, residents who depend upon the groundwater sources should use boiled water before drinking. However, it is suitable for other domestic needs.

The local authorities should regularly maintain groundwater monitoring program to check the spreading of contaminants before and after the implementation of on-site sanitation system.

On-site sanitation system implementation site should take into account the local hydrogeological conditions before implementation. The study becomes relevant in hard rock settings wherein the movement of contaminants is increased in the event of presence of potential pathways.